06_cleanup_linkres.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Linkres cleanup and filtering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Boilerplate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/mnt/home/pleung/projects/bistable_bundle/r4/hinges\n",
      "dig80\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 1;\n",
       "                var nbb_unformatted_code = \"%load_ext nb_black\\n# python internal\\nimport collections\\nimport copy\\nimport gc\\nfrom glob import glob\\nimport h5py\\nimport itertools\\nimport os\\nimport random\\nimport re\\nimport socket\\nimport shutil\\nimport subprocess\\nimport sys\\n\\n# conda/pip\\nimport dask\\nimport graphviz\\nimport matplotlib.pyplot as plt\\nimport pandas as pd\\nimport numpy as np\\nimport scipy\\nimport seaborn as sns\\nfrom tqdm import tqdm\\n\\n# special packages on the DIGS\\nimport py3Dmol\\nimport pymol\\nimport pyrosetta\\n\\n# notebook magic\\n%matplotlib inline\\n%load_ext autoreload\\n%autoreload 2\\n\\nprint(os.getcwd())\\nprint(socket.gethostname())\";\n",
       "                var nbb_formatted_code = \"%load_ext nb_black\\n# python internal\\nimport collections\\nimport copy\\nimport gc\\nfrom glob import glob\\nimport h5py\\nimport itertools\\nimport os\\nimport random\\nimport re\\nimport socket\\nimport shutil\\nimport subprocess\\nimport sys\\n\\n# conda/pip\\nimport dask\\nimport graphviz\\nimport matplotlib.pyplot as plt\\nimport pandas as pd\\nimport numpy as np\\nimport scipy\\nimport seaborn as sns\\nfrom tqdm import tqdm\\n\\n# special packages on the DIGS\\nimport py3Dmol\\nimport pymol\\nimport pyrosetta\\n\\n# notebook magic\\n%matplotlib inline\\n%load_ext autoreload\\n%autoreload 2\\n\\nprint(os.getcwd())\\nprint(socket.gethostname())\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load_ext nb_black\n",
    "# python internal\n",
    "import collections\n",
    "import copy\n",
    "import gc\n",
    "from glob import glob\n",
    "import h5py\n",
    "import itertools\n",
    "import os\n",
    "import random\n",
    "import re\n",
    "import socket\n",
    "import shutil\n",
    "import subprocess\n",
    "import sys\n",
    "\n",
    "# conda/pip\n",
    "import dask\n",
    "import graphviz\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy\n",
    "import seaborn as sns\n",
    "from tqdm import tqdm\n",
    "\n",
    "# special packages on the DIGS\n",
    "import py3Dmol\n",
    "import pymol\n",
    "import pyrosetta\n",
    "\n",
    "# notebook magic\n",
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "print(os.getcwd())\n",
    "print(socket.gethostname())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate state X and state Y with and without disulfides by doing `KeepSequenceSymmetry` and fixbb design\n",
    "I will use the serialization build of PyRosetta to enable recording user defined info about the designs.  \n",
    "This enables downstream inline filtering and data analysis, as well as clustering by lineage.\n",
    "\n",
    "Record rich information about residues that might be interesting to show/direct switching, GLY, HIS and TRP.  \n",
    "Record info about sequences and save pdbs for X and Y, with and without disulfide staples  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 2;\n",
       "                var nbb_unformatted_code = \"from pyrosetta.distributed.packed_pose.core import PackedPose\\n\\n\\ndef almost_linkres(packed_pose_in=None, **kwargs) -> PackedPose:\\n    \\\"\\\"\\\"\\n    Load pose. Design with linkres x3: state_X_nocys, state_Y_nocys,\\n    state_XY_nocys. Dump X and Y poses with and without disulfides (4 total).\\n    Return XY without the disulfides.\\n    \\\"\\\"\\\"\\n    import os\\n    import pyrosetta\\n    import pyrosetta.distributed.io as io\\n    from pyrosetta.distributed import cluster\\n    from pyrosetta.rosetta.core.pack.task.operation import (\\n        IncludeCurrent,\\n        InitializeFromCommandline,\\n        ExtraRotamersGeneric,\\n    )\\n    from pyrosetta.rosetta.protocols.denovo_design.movers import FastDesign\\n    from pyrosetta.rosetta.protocols.rosetta_scripts import XmlObjects\\n    from pyrosetta.rosetta.protocols.task_operations import LimitAromaChi2Operation\\n\\n    def parse_scoreinfo(pdb_path) -> dict:\\n        \\\"\\\"\\\"\\n        Hacky unsafe way to load scoreinfo written to pdb by rosetta.\\n        Probably should add additional pdb strings to skip\\n        \\\"\\\"\\\"\\n        with open(pdb_path, \\\"r\\\") as f:\\n            scores = {}\\n            pdb_strings = [\\\"HEADER\\\", \\\"EXPDTA\\\", \\\"SSBOND\\\", \\\"ATOM\\\", \\\"TER\\\", \\\"REMARK\\\", \\\"#\\\"]\\n            for line in f:\\n                if line == \\\"\\\\n\\\":  # skip empty lines\\n                    pass\\n                # skip pdb specific rows\\n                elif any(pdb_string in line for pdb_string in pdb_strings):\\n                    pass\\n                else:\\n                    key_value = line.rstrip().split()\\n                    # handle pose energies table\\n                    try:\\n                        assert len(key_value) == 2\\n                    except AssertionError:  # pose energies have many values per (residue) key\\n                        continue\\n                    key, value = tuple(key_value)\\n                    if any(c.isalpha() for c in value):  # handle as str below\\n                        pass\\n                    elif \\\",\\\" in value:  # handle as str below\\n                        pass\\n                    elif \\\".\\\" in value:  # if it has a . it maybe a float\\n                        try:\\n                            value = float(value)\\n                        except ValueError:\\n                            pass\\n                    else:  # if all else fails maybe an int\\n                        try:\\n                            value = int(value)\\n                        except ValueError:\\n                            pass\\n                    scores[key] = value  # if all typecasting failed it is still a str\\n        return scores\\n\\n    def score(packed_pose_in: PackedPose, **kwargs) -> PackedPose:\\n        \\\"\\\"\\\"\\n        Sequence\\n        Compute HIS SASA for X, Y, (compute delta on df)\\n        Compute buried HIS (probably in interface)\\n        Count GLY near interface, (compute glycine count on df)\\n        Compute TRP SASA for X, Y, (compute delta on df)\\n        Count TRP nearby aromatic non-TRP neighbors\\n        Assumes middle is where the switch is split originally, enabling forward and backward SSElement selection approach\\n        @pleung\\n        \\\"\\\"\\\"\\n        import pyrosetta\\n        import pyrosetta.distributed.io as io\\n        from pyrosetta.distributed.tasks.rosetta_scripts import (\\n            SingleoutputRosettaScriptsTask,\\n        )\\n\\n        scores = kwargs[\\\"score_dict\\\"]\\n        xml = \\\"\\\"\\\"\\n        <ROSETTASCRIPTS>\\n            <SCOREFXNS>\\n            </SCOREFXNS>\\n            <RESIDUE_SELECTORS>\\n                <ResidueName name=\\\"GLY\\\" residue_name3=\\\"GLY\\\" />\\n                <ResidueName name=\\\"HIS\\\" residue_name3=\\\"HIS\\\" />\\n                <ResidueName name=\\\"TRP\\\" residue_name3=\\\"TRP\\\" />\\n                <ResidueName name=\\\"aro\\\" residue_name3=\\\"TYR,PHE,HIS\\\" />\\n                <Layer name=\\\"surface\\\" select_core=\\\"false\\\" select_boundary=\\\"false\\\" select_surface=\\\"true\\\"\\n                    use_sidechain_neighbors=\\\"true\\\"/>\\n                <Layer name=\\\"boundary\\\" select_core=\\\"false\\\" select_boundary=\\\"true\\\" select_surface=\\\"false\\\" \\n                    use_sidechain_neighbors=\\\"true\\\"/>\\n                <Layer name=\\\"core\\\" select_core=\\\"true\\\" select_boundary=\\\"false\\\" select_surface=\\\"false\\\" \\n                    use_sidechain_neighbors=\\\"true\\\"/>\\n                <SecondaryStructure name=\\\"sheet\\\" overlap=\\\"0\\\" minH=\\\"3\\\" minE=\\\"2\\\" include_terminal_loops=\\\"false\\\" \\n                    use_dssp=\\\"true\\\" ss=\\\"E\\\"/>\\n                <SecondaryStructure name=\\\"entire_loop\\\" overlap=\\\"0\\\" minH=\\\"3\\\" minE=\\\"2\\\" include_terminal_loops=\\\"true\\\" \\n                    use_dssp=\\\"true\\\" ss=\\\"L\\\"/>\\n                <SecondaryStructure name=\\\"entire_helix\\\" overlap=\\\"0\\\" minH=\\\"3\\\" minE=\\\"2\\\" include_terminal_loops=\\\"false\\\"\\n                    use_dssp=\\\"true\\\" ss=\\\"H\\\"/>\\n                <And name=\\\"helix_cap\\\" selectors=\\\"entire_loop\\\">\\n                    <PrimarySequenceNeighborhood lower=\\\"1\\\" upper=\\\"0\\\" selector=\\\"entire_helix\\\"/>\\n                </And>\\n                <And name=\\\"helix_start\\\" selectors=\\\"entire_helix\\\">\\n                    <PrimarySequenceNeighborhood lower=\\\"0\\\" upper=\\\"1\\\" selector=\\\"helix_cap\\\"/>\\n                </And>\\n                <And name=\\\"helix\\\" selectors=\\\"entire_helix\\\">\\n                    <Not selector=\\\"helix_start\\\"/>\\n                </And>\\n                <And name=\\\"loop\\\" selectors=\\\"entire_loop\\\">\\n                    <Not selector=\\\"helix_cap\\\"/>\\n                </And>\\n                <SSElement name=\\\"near_int\\\" selection=\\\"{pre},H,S\\\" to_selection=\\\"-{post},H,E\\\" chain=\\\"A\\\" reassign_short_terminal_loop=\\\"2\\\" />            \\n                <Neighborhood name=\\\"near_TRP\\\" selector=\\\"TRP\\\" distance=\\\"6.0\\\" />\\n                <And name=\\\"GLY_near_int\\\" selectors=\\\"near_int,HIS\\\" />\\n                <And name=\\\"core_HIS\\\" selectors=\\\"core,HIS\\\" />\\n                <And name=\\\"aro_near_TRP\\\" selectors=\\\"aro,near_TRP\\\" />\\n            </RESIDUE_SELECTORS>\\n            <TASKOPERATIONS>\\n            </TASKOPERATIONS>\\n            <FILTERS>\\n                <BuriedSurfaceArea name=\\\"HIS_burial\\\" residue_selector=\\\"HIS\\\" confidence=\\\"0\\\" />\\n                <BuriedSurfaceArea name=\\\"TRP_burial\\\" residue_selector=\\\"TRP\\\" confidence=\\\"0\\\" />\\n                <ResidueCount name=\\\"buried_HIS\\\" residue_selector=\\\"core_HIS\\\" />\\n                <ResidueCount name=\\\"near_int_GLY\\\" residue_selector=\\\"GLY_near_int\\\" />\\n                <ResidueCount name=\\\"near_TRP_aro\\\" residue_selector=\\\"aro_near_TRP\\\" />\\n            </FILTERS>\\n            <MOVERS>\\n            </MOVERS>\\n            <PROTOCOLS>\\n                <Add filter=\\\"HIS_burial\\\" />\\n                <Add filter=\\\"TRP_burial\\\" />\\n                <Add filter=\\\"buried_HIS\\\" />\\n                <Add filter=\\\"near_int_GLY\\\" />\\n                <Add filter=\\\"near_TRP_aro\\\" />\\n            </PROTOCOLS>\\n        </ROSETTASCRIPTS>\\n        \\\"\\\"\\\".format(\\n            pre=int(int(scores[\\\"pre_break_helix\\\"]) - 1),\\n            post=int(int(scores[\\\"pre_break_helix\\\"]) - 1),\\n        )\\n        # relax and score pose\\n        score = SingleoutputRosettaScriptsTask(xml)\\n        scored_ppose = score(packed_pose_in.pose.clone())\\n        update = scored_ppose.pose.scores\\n        pose = io.to_pose(scored_ppose.pose.clone())\\n        # update scores for values that were updated\\n        for key, value in scores.items():\\n            if key in update.keys():\\n                pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, update[key])\\n            else:\\n                pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, value)\\n        final_ppose = io.to_packed(pose)\\n        return final_ppose\\n\\n    if packed_pose_in == None:\\n        pdb_path = kwargs[\\\"-s\\\"]\\n    else:\\n        raise RuntimeError(\\n            \\\"This method requires a pdb path, it cannot accept a PackedPose as input\\\"\\n        )\\n    scores = parse_scoreinfo(pdb_path)\\n    ppose = io.pose_from_file(pdb_path)\\n    pose = io.to_pose(ppose)\\n    if kwargs[\\\"out_path\\\"] == None:\\n        out_path = os.getcwd()\\n    else:\\n        out_path = kwargs[\\\"out_path\\\"]\\n        os.makedirs(os.path.join(os.getcwd(), out_path), exist_ok=True)\\n    cys_X = scores[\\\"disulfide_at_X\\\"].split(\\\",\\\")\\n    cys_Y = scores[\\\"disulfide_at_Y\\\"].split(\\\",\\\")\\n    unique_cys = set(cys_X + cys_Y)\\n    # check whether disulfides are orthogonal\\n    design_resis = [int(x) for x in unique_cys]\\n    try:\\n        assert len(unique_cys) in [\\n            3,\\n            4,\\n        ]  # if there are less than 3 CYS or more than 4 we really f'd up at some point\\n    except AssertionError:\\n        print(\\\"There should be at least 3 unique CYS, but no more than 4\\\")\\n        return None\\n    cys1_X, cys2_X = int(cys_X[0]), int(cys_X[1])\\n    cys1_Y, cys2_Y = int(cys_Y[0]), int(cys_Y[1])\\n    length = scores[\\\"total_length\\\"]\\n    resis_matrix = [[x, int(x + length)] for x in design_resis]\\n    target_resis = \\\",\\\".join(\\n        [str(x) for row in resis_matrix for x in row]\\n    )  # flatten matrix\\n    xml_string = \\\"\\\"\\\"\\n    <ROSETTASCRIPTS>\\n        <SCOREFXNS>\\n            <ScoreFunction name=\\\"sfxn_design\\\" weights=\\\"beta_nov16\\\" >\\n                <Set use_hb_env_dep=\\\"true\\\" />\\n                <Reweight scoretype=\\\"approximate_buried_unsat_penalty\\\" weight=\\\"17\\\" />\\n                <Set approximate_buried_unsat_penalty_burial_atomic_depth=\\\"3.5\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_energy_threshold=\\\"-1.0\\\" />\\n                <Set approximate_buried_unsat_penalty_natural_corrections1=\\\"true\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_cross_chain=\\\"-7\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_ser_to_helix_bb=\\\"1\\\"/>                    \\n            </ScoreFunction>\\n        </SCOREFXNS>\\n        <RESIDUE_SELECTORS>\\n            <Chain name=\\\"chA\\\" chains=\\\"A\\\" />\\n            <Chain name=\\\"chB\\\" chains=\\\"B\\\" />\\n            <Index name=\\\"designable\\\" resnums=\\\"{target_resis}\\\" />\\n            <Neighborhood name=\\\"packable\\\" selector=\\\"designable\\\" distance=\\\"6.0\\\" include_focus_in_subset=\\\"true\\\" />\\n            <Not name=\\\"not_designable\\\" selector=\\\"designable\\\" />\\n            <Not name=\\\"not_packable\\\" selector=\\\"packable\\\" />\\n        </RESIDUE_SELECTORS>\\n        <TASKOPERATIONS>\\n            <OperateOnResidueSubset name=\\\"design\\\" selector=\\\"designable\\\"> # no CYS; GLY; PRO;\\n                <RestrictAbsentCanonicalAASRLT aas=\\\"ADEFHIKLMNQRSTVWY\\\" />\\n            </OperateOnResidueSubset>\\n            <OperateOnResidueSubset name=\\\"pack\\\" selector=\\\"not_designable\\\">\\n                <RestrictToRepackingRLT/>\\n            </OperateOnResidueSubset>\\n            <OperateOnResidueSubset name=\\\"lock\\\" selector=\\\"not_packable\\\">\\n                <PreventRepackingRLT/>\\n            </OperateOnResidueSubset>\\n            <RestrictToRepacking name=\\\"rtrp\\\" />\\n            <KeepSequenceSymmetry name=\\\"2state\\\" setting=\\\"true\\\"/>\\n        </TASKOPERATIONS>\\n        <MOVERS>\\n            <SetupForSequenceSymmetryMover name=\\\"almost_linkres\\\" sequence_symmetry_behaviour=\\\"2state\\\" >\\n                <SequenceSymmetry residue_selectors=\\\"chA,chB\\\" />\\n            </SetupForSequenceSymmetryMover>\\n            <SwitchChainOrder name=\\\"delete_Y\\\" chain_order=\\\"1\\\"/>\\n            <SwitchChainOrder name=\\\"delete_X\\\" chain_order=\\\"2\\\"/>\\n            <MutateResidue name=\\\"cys1_X\\\" target=\\\"{cys1_X}\\\" new_res=\\\"CYS\\\" />\\n            <MutateResidue name=\\\"cys2_X\\\" target=\\\"{cys2_X}\\\" new_res=\\\"CYS\\\" />\\n            <MutateResidue name=\\\"cys1_Y\\\" target=\\\"{cys1_Y}\\\" new_res=\\\"CYS\\\" />\\n            <MutateResidue name=\\\"cys2_Y\\\" target=\\\"{cys2_Y}\\\" new_res=\\\"CYS\\\" />\\n            <ForceDisulfides name=\\\"restore_X\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                disulfides=\\\"{cys1_X}:{cys2_X}\\\"\\n                remove_existing=\\\"false\\\"\\n                repack=\\\"true\\\" />\\n            <ForceDisulfides name=\\\"restore_Y\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                disulfides=\\\"{cys1_Y}:{cys2_Y}\\\"\\n                remove_existing=\\\"false\\\"\\n                repack=\\\"true\\\" />\\n            <FastRelax name=\\\"relax\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                repeats=\\\"1\\\" \\n                relaxscript=\\\"MonomerRelax2019\\\"\\n                />\\n            <PackRotamersMover name=\\\"repack\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                task_operations=\\\"rtrp\\\"\\n                />\\n        </MOVERS>\\n        <FILTERS>\\n        </FILTERS>\\n        <SIMPLE_METRICS>\\n        </SIMPLE_METRICS>\\n        <APPLY_TO_POSE>\\n        </APPLY_TO_POSE>\\n        <PROTOCOLS>\\n        </PROTOCOLS>\\n    </ROSETTASCRIPTS>\\n    \\\"\\\"\\\".format(\\n        target_resis=target_resis,\\n        cys1_X=cys1_X,\\n        cys2_X=cys2_X,\\n        cys1_Y=cys1_Y,\\n        cys2_Y=cys2_Y,\\n    )\\n    # set taskops by extracting them from the xml_string\\n    xml_obj = XmlObjects.create_from_string(xml_string)\\n    design_task = xml_obj.get_task_operation(\\\"design\\\")\\n    pack_task = xml_obj.get_task_operation(\\\"pack\\\")\\n    lock_task = xml_obj.get_task_operation(\\\"lock\\\")\\n    linkres_task = xml_obj.get_task_operation(\\\"2state\\\")\\n    # set taskops\\n    task_factory = pyrosetta.rosetta.core.pack.task.TaskFactory()\\n    task_factory.push_back(IncludeCurrent())\\n    task_factory.push_back(InitializeFromCommandline())\\n    arochi = LimitAromaChi2Operation()\\n    arochi.include_trp(True)\\n    task_factory.push_back(arochi)\\n    ex1_ex2 = ExtraRotamersGeneric()\\n    ex1_ex2.ex1(True)\\n    ex1_ex2.ex2(True)\\n    task_factory.push_back(ex1_ex2)\\n    task_factory.push_back(design_task)\\n    task_factory.push_back(pack_task)\\n    task_factory.push_back(lock_task)\\n    task_factory.push_back(linkres_task)\\n    # set movemap\\n    mm = pyrosetta.rosetta.core.kinematics.MoveMap()\\n    mm.set_bb(False)\\n    mm.set_chi(True)\\n    mm.set_jump(False)\\n    sfxn_design = xml_obj.get_score_function(\\\"sfxn_design\\\")\\n    fast_design = FastDesign(scorefxn_in=sfxn_design, standard_repeats=1)\\n    fast_design.cartesian(False)\\n    fast_design.set_task_factory(task_factory)\\n    fast_design.set_movemap(mm)\\n    fast_design.minimize_bond_angles(False)\\n    fast_design.minimize_bond_lengths(False)\\n    fast_design.min_type(\\\"lbfgs_armijo_nonmonotone\\\")\\n    fast_design.ramp_down_constraints(False)\\n\\n    name = pose.pdb_info().name()\\n    basename_no_ext = name.split(\\\"/\\\")[-1].replace(\\\".pdb\\\", \\\"\\\", 1)\\n    if out_path[-1] != \\\"/\\\":\\n        out_path += \\\"/\\\"\\n    else:\\n        pass\\n    pose = pose.clone()\\n    almost_linkres = xml_obj.get_mover(\\\"almost_linkres\\\")\\n    almost_linkres.apply(pose)\\n    fast_design.apply(pose)\\n\\n    ref_pose = pose.clone()\\n    # get state X\\n    delete_Y = xml_obj.get_mover(\\\"delete_Y\\\")\\n    delete_Y.apply(pose)\\n    ref_X = pose.clone()\\n    cys1 = xml_obj.get_mover(\\\"cys1_X\\\")\\n    cys2 = xml_obj.get_mover(\\\"cys2_X\\\")\\n    restore = xml_obj.get_mover(\\\"restore_X\\\")\\n    relax = xml_obj.get_mover(\\\"relax\\\")\\n    repack = xml_obj.get_mover(\\\"repack\\\")\\n    cys1.apply(pose)\\n    cys2.apply(pose)\\n    restore.apply(pose)\\n    repack.apply(pose)\\n    # dump X\\n    pose.dump_pdb(os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_X.pdb\\\"))\\n    scores[\\\"X_seq\\\"] = pose.sequence()\\n    # get rmsd\\n    relax_X = ref_X.clone()\\n    relax.apply(relax_X)\\n    # dump X_nocys\\n    relax_X.dump_pdb(\\n        os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_X_nocys.pdb\\\")\\n    )\\n    rmsd_calc = pyrosetta.rosetta.core.simple_metrics.metrics.RMSDMetric()\\n    rmsd_calc.set_rmsd_type(pyrosetta.rosetta.core.scoring.rmsd_atoms(3))\\n    rmsd_calc.set_run_superimpose(True)\\n    rmsd_calc.set_comparison_pose(ref_X)\\n    rmsd = rmsd_calc.calculate(relax_X)\\n    scores[\\\"rmsd_final_X\\\"] = rmsd\\n    # get state Y\\n    pose = ref_pose.clone()\\n    delete_X = xml_obj.get_mover(\\\"delete_X\\\")\\n    delete_X.apply(pose)\\n    ref_Y = pose.clone()\\n    cys1 = xml_obj.get_mover(\\\"cys1_Y\\\")\\n    cys2 = xml_obj.get_mover(\\\"cys2_Y\\\")\\n    restore = xml_obj.get_mover(\\\"restore_Y\\\")\\n    cys1.apply(pose)\\n    cys2.apply(pose)\\n    restore.apply(pose)\\n    repack.apply(pose)\\n    # dump Y\\n    pose.dump_pdb(os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_Y.pdb\\\"))\\n    scores[\\\"Y_seq\\\"] = pose.sequence()\\n    # get rmsd\\n    relax_Y = ref_Y.clone()\\n    relax.apply(relax_Y)\\n    # dump Y_nocys\\n    relax_Y.dump_pdb(\\n        os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_Y_nocys.pdb\\\")\\n    )\\n    rmsd_calc = pyrosetta.rosetta.core.simple_metrics.metrics.RMSDMetric()\\n    rmsd_calc.set_rmsd_type(pyrosetta.rosetta.core.scoring.rmsd_atoms(3))\\n    rmsd_calc.set_run_superimpose(True)\\n    rmsd_calc.set_comparison_pose(ref_Y)\\n    rmsd = rmsd_calc.calculate(relax_Y)\\n    scores[\\\"rmsd_final_Y\\\"] = rmsd\\n    assert relax_X.sequence() == relax_Y.sequence()\\n    scores[\\\"final_sequence\\\"] = relax_X.sequence()\\n    # put in paths to X, Y, X_nocys, Y_nocys\\n    for suffix in [\\\"X\\\", \\\"Y\\\", \\\"X_nocys\\\", \\\"Y_nocys\\\"]:\\n        scores[f\\\"{suffix}_path\\\"] = os.path.join(\\n            os.getcwd(), out_path + f\\\"{basename_no_ext}_{suffix}.pdb\\\"\\n        )\\n    # score relax_X and relax_Y, update scores with those scores\\n    scored_X = score(io.to_packed(relax_X), score_dict=scores)\\n    scored_Y = score(io.to_packed(relax_Y), score_dict=scores)\\n    updates = []\\n    keys = [\\\"HIS_burial\\\", \\\"TRP_burial\\\", \\\"buried_HIS\\\", \\\"near_int_GLY\\\", \\\"near_TRP_aro\\\"]\\n\\n    for chain, score_dict in [\\n        (\\\"X\\\", dict(scored_X.pose.scores)),\\n        (\\\"Y\\\", dict(scored_Y.pose.scores)),\\n    ]:\\n        updates.append(\\n            {\\n                f\\\"{key}_{chain}\\\": value\\n                for key, value in score_dict.items()\\n                if key in keys\\n            }\\n        )\\n    scores.update({**updates[0], **updates[1]})\\n    for key, value in scores.items():\\n        pyrosetta.rosetta.core.pose.setPoseExtraScore(ref_pose, key, value)\\n    ppose = io.to_packed(ref_pose)\\n    return ppose\";\n",
       "                var nbb_formatted_code = \"from pyrosetta.distributed.packed_pose.core import PackedPose\\n\\n\\ndef almost_linkres(packed_pose_in=None, **kwargs) -> PackedPose:\\n    \\\"\\\"\\\"\\n    Load pose. Design with linkres x3: state_X_nocys, state_Y_nocys,\\n    state_XY_nocys. Dump X and Y poses with and without disulfides (4 total).\\n    Return XY without the disulfides.\\n    \\\"\\\"\\\"\\n    import os\\n    import pyrosetta\\n    import pyrosetta.distributed.io as io\\n    from pyrosetta.distributed import cluster\\n    from pyrosetta.rosetta.core.pack.task.operation import (\\n        IncludeCurrent,\\n        InitializeFromCommandline,\\n        ExtraRotamersGeneric,\\n    )\\n    from pyrosetta.rosetta.protocols.denovo_design.movers import FastDesign\\n    from pyrosetta.rosetta.protocols.rosetta_scripts import XmlObjects\\n    from pyrosetta.rosetta.protocols.task_operations import LimitAromaChi2Operation\\n\\n    def parse_scoreinfo(pdb_path) -> dict:\\n        \\\"\\\"\\\"\\n        Hacky unsafe way to load scoreinfo written to pdb by rosetta.\\n        Probably should add additional pdb strings to skip\\n        \\\"\\\"\\\"\\n        with open(pdb_path, \\\"r\\\") as f:\\n            scores = {}\\n            pdb_strings = [\\\"HEADER\\\", \\\"EXPDTA\\\", \\\"SSBOND\\\", \\\"ATOM\\\", \\\"TER\\\", \\\"REMARK\\\", \\\"#\\\"]\\n            for line in f:\\n                if line == \\\"\\\\n\\\":  # skip empty lines\\n                    pass\\n                # skip pdb specific rows\\n                elif any(pdb_string in line for pdb_string in pdb_strings):\\n                    pass\\n                else:\\n                    key_value = line.rstrip().split()\\n                    # handle pose energies table\\n                    try:\\n                        assert len(key_value) == 2\\n                    except AssertionError:  # pose energies have many values per (residue) key\\n                        continue\\n                    key, value = tuple(key_value)\\n                    if any(c.isalpha() for c in value):  # handle as str below\\n                        pass\\n                    elif \\\",\\\" in value:  # handle as str below\\n                        pass\\n                    elif \\\".\\\" in value:  # if it has a . it maybe a float\\n                        try:\\n                            value = float(value)\\n                        except ValueError:\\n                            pass\\n                    else:  # if all else fails maybe an int\\n                        try:\\n                            value = int(value)\\n                        except ValueError:\\n                            pass\\n                    scores[key] = value  # if all typecasting failed it is still a str\\n        return scores\\n\\n    def score(packed_pose_in: PackedPose, **kwargs) -> PackedPose:\\n        \\\"\\\"\\\"\\n        Sequence\\n        Compute HIS SASA for X, Y, (compute delta on df)\\n        Compute buried HIS (probably in interface)\\n        Count GLY near interface, (compute glycine count on df)\\n        Compute TRP SASA for X, Y, (compute delta on df)\\n        Count TRP nearby aromatic non-TRP neighbors\\n        Assumes middle is where the switch is split originally, enabling forward and backward SSElement selection approach\\n        @pleung\\n        \\\"\\\"\\\"\\n        import pyrosetta\\n        import pyrosetta.distributed.io as io\\n        from pyrosetta.distributed.tasks.rosetta_scripts import (\\n            SingleoutputRosettaScriptsTask,\\n        )\\n\\n        scores = kwargs[\\\"score_dict\\\"]\\n        xml = \\\"\\\"\\\"\\n        <ROSETTASCRIPTS>\\n            <SCOREFXNS>\\n            </SCOREFXNS>\\n            <RESIDUE_SELECTORS>\\n                <ResidueName name=\\\"GLY\\\" residue_name3=\\\"GLY\\\" />\\n                <ResidueName name=\\\"HIS\\\" residue_name3=\\\"HIS\\\" />\\n                <ResidueName name=\\\"TRP\\\" residue_name3=\\\"TRP\\\" />\\n                <ResidueName name=\\\"aro\\\" residue_name3=\\\"TYR,PHE,HIS\\\" />\\n                <Layer name=\\\"surface\\\" select_core=\\\"false\\\" select_boundary=\\\"false\\\" select_surface=\\\"true\\\"\\n                    use_sidechain_neighbors=\\\"true\\\"/>\\n                <Layer name=\\\"boundary\\\" select_core=\\\"false\\\" select_boundary=\\\"true\\\" select_surface=\\\"false\\\" \\n                    use_sidechain_neighbors=\\\"true\\\"/>\\n                <Layer name=\\\"core\\\" select_core=\\\"true\\\" select_boundary=\\\"false\\\" select_surface=\\\"false\\\" \\n                    use_sidechain_neighbors=\\\"true\\\"/>\\n                <SecondaryStructure name=\\\"sheet\\\" overlap=\\\"0\\\" minH=\\\"3\\\" minE=\\\"2\\\" include_terminal_loops=\\\"false\\\" \\n                    use_dssp=\\\"true\\\" ss=\\\"E\\\"/>\\n                <SecondaryStructure name=\\\"entire_loop\\\" overlap=\\\"0\\\" minH=\\\"3\\\" minE=\\\"2\\\" include_terminal_loops=\\\"true\\\" \\n                    use_dssp=\\\"true\\\" ss=\\\"L\\\"/>\\n                <SecondaryStructure name=\\\"entire_helix\\\" overlap=\\\"0\\\" minH=\\\"3\\\" minE=\\\"2\\\" include_terminal_loops=\\\"false\\\"\\n                    use_dssp=\\\"true\\\" ss=\\\"H\\\"/>\\n                <And name=\\\"helix_cap\\\" selectors=\\\"entire_loop\\\">\\n                    <PrimarySequenceNeighborhood lower=\\\"1\\\" upper=\\\"0\\\" selector=\\\"entire_helix\\\"/>\\n                </And>\\n                <And name=\\\"helix_start\\\" selectors=\\\"entire_helix\\\">\\n                    <PrimarySequenceNeighborhood lower=\\\"0\\\" upper=\\\"1\\\" selector=\\\"helix_cap\\\"/>\\n                </And>\\n                <And name=\\\"helix\\\" selectors=\\\"entire_helix\\\">\\n                    <Not selector=\\\"helix_start\\\"/>\\n                </And>\\n                <And name=\\\"loop\\\" selectors=\\\"entire_loop\\\">\\n                    <Not selector=\\\"helix_cap\\\"/>\\n                </And>\\n                <SSElement name=\\\"near_int\\\" selection=\\\"{pre},H,S\\\" to_selection=\\\"-{post},H,E\\\" chain=\\\"A\\\" reassign_short_terminal_loop=\\\"2\\\" />            \\n                <Neighborhood name=\\\"near_TRP\\\" selector=\\\"TRP\\\" distance=\\\"6.0\\\" />\\n                <And name=\\\"GLY_near_int\\\" selectors=\\\"near_int,HIS\\\" />\\n                <And name=\\\"core_HIS\\\" selectors=\\\"core,HIS\\\" />\\n                <And name=\\\"aro_near_TRP\\\" selectors=\\\"aro,near_TRP\\\" />\\n            </RESIDUE_SELECTORS>\\n            <TASKOPERATIONS>\\n            </TASKOPERATIONS>\\n            <FILTERS>\\n                <BuriedSurfaceArea name=\\\"HIS_burial\\\" residue_selector=\\\"HIS\\\" confidence=\\\"0\\\" />\\n                <BuriedSurfaceArea name=\\\"TRP_burial\\\" residue_selector=\\\"TRP\\\" confidence=\\\"0\\\" />\\n                <ResidueCount name=\\\"buried_HIS\\\" residue_selector=\\\"core_HIS\\\" />\\n                <ResidueCount name=\\\"near_int_GLY\\\" residue_selector=\\\"GLY_near_int\\\" />\\n                <ResidueCount name=\\\"near_TRP_aro\\\" residue_selector=\\\"aro_near_TRP\\\" />\\n            </FILTERS>\\n            <MOVERS>\\n            </MOVERS>\\n            <PROTOCOLS>\\n                <Add filter=\\\"HIS_burial\\\" />\\n                <Add filter=\\\"TRP_burial\\\" />\\n                <Add filter=\\\"buried_HIS\\\" />\\n                <Add filter=\\\"near_int_GLY\\\" />\\n                <Add filter=\\\"near_TRP_aro\\\" />\\n            </PROTOCOLS>\\n        </ROSETTASCRIPTS>\\n        \\\"\\\"\\\".format(\\n            pre=int(int(scores[\\\"pre_break_helix\\\"]) - 1),\\n            post=int(int(scores[\\\"pre_break_helix\\\"]) - 1),\\n        )\\n        # relax and score pose\\n        score = SingleoutputRosettaScriptsTask(xml)\\n        scored_ppose = score(packed_pose_in.pose.clone())\\n        update = scored_ppose.pose.scores\\n        pose = io.to_pose(scored_ppose.pose.clone())\\n        # update scores for values that were updated\\n        for key, value in scores.items():\\n            if key in update.keys():\\n                pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, update[key])\\n            else:\\n                pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, value)\\n        final_ppose = io.to_packed(pose)\\n        return final_ppose\\n\\n    if packed_pose_in == None:\\n        pdb_path = kwargs[\\\"-s\\\"]\\n    else:\\n        raise RuntimeError(\\n            \\\"This method requires a pdb path, it cannot accept a PackedPose as input\\\"\\n        )\\n    scores = parse_scoreinfo(pdb_path)\\n    ppose = io.pose_from_file(pdb_path)\\n    pose = io.to_pose(ppose)\\n    if kwargs[\\\"out_path\\\"] == None:\\n        out_path = os.getcwd()\\n    else:\\n        out_path = kwargs[\\\"out_path\\\"]\\n        os.makedirs(os.path.join(os.getcwd(), out_path), exist_ok=True)\\n    cys_X = scores[\\\"disulfide_at_X\\\"].split(\\\",\\\")\\n    cys_Y = scores[\\\"disulfide_at_Y\\\"].split(\\\",\\\")\\n    unique_cys = set(cys_X + cys_Y)\\n    # check whether disulfides are orthogonal\\n    design_resis = [int(x) for x in unique_cys]\\n    try:\\n        assert len(unique_cys) in [\\n            3,\\n            4,\\n        ]  # if there are less than 3 CYS or more than 4 we really f'd up at some point\\n    except AssertionError:\\n        print(\\\"There should be at least 3 unique CYS, but no more than 4\\\")\\n        return None\\n    cys1_X, cys2_X = int(cys_X[0]), int(cys_X[1])\\n    cys1_Y, cys2_Y = int(cys_Y[0]), int(cys_Y[1])\\n    length = scores[\\\"total_length\\\"]\\n    resis_matrix = [[x, int(x + length)] for x in design_resis]\\n    target_resis = \\\",\\\".join(\\n        [str(x) for row in resis_matrix for x in row]\\n    )  # flatten matrix\\n    xml_string = \\\"\\\"\\\"\\n    <ROSETTASCRIPTS>\\n        <SCOREFXNS>\\n            <ScoreFunction name=\\\"sfxn_design\\\" weights=\\\"beta_nov16\\\" >\\n                <Set use_hb_env_dep=\\\"true\\\" />\\n                <Reweight scoretype=\\\"approximate_buried_unsat_penalty\\\" weight=\\\"17\\\" />\\n                <Set approximate_buried_unsat_penalty_burial_atomic_depth=\\\"3.5\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_energy_threshold=\\\"-1.0\\\" />\\n                <Set approximate_buried_unsat_penalty_natural_corrections1=\\\"true\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_cross_chain=\\\"-7\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_ser_to_helix_bb=\\\"1\\\"/>                    \\n            </ScoreFunction>\\n        </SCOREFXNS>\\n        <RESIDUE_SELECTORS>\\n            <Chain name=\\\"chA\\\" chains=\\\"A\\\" />\\n            <Chain name=\\\"chB\\\" chains=\\\"B\\\" />\\n            <Index name=\\\"designable\\\" resnums=\\\"{target_resis}\\\" />\\n            <Neighborhood name=\\\"packable\\\" selector=\\\"designable\\\" distance=\\\"6.0\\\" include_focus_in_subset=\\\"true\\\" />\\n            <Not name=\\\"not_designable\\\" selector=\\\"designable\\\" />\\n            <Not name=\\\"not_packable\\\" selector=\\\"packable\\\" />\\n        </RESIDUE_SELECTORS>\\n        <TASKOPERATIONS>\\n            <OperateOnResidueSubset name=\\\"design\\\" selector=\\\"designable\\\"> # no CYS; GLY; PRO;\\n                <RestrictAbsentCanonicalAASRLT aas=\\\"ADEFHIKLMNQRSTVWY\\\" />\\n            </OperateOnResidueSubset>\\n            <OperateOnResidueSubset name=\\\"pack\\\" selector=\\\"not_designable\\\">\\n                <RestrictToRepackingRLT/>\\n            </OperateOnResidueSubset>\\n            <OperateOnResidueSubset name=\\\"lock\\\" selector=\\\"not_packable\\\">\\n                <PreventRepackingRLT/>\\n            </OperateOnResidueSubset>\\n            <RestrictToRepacking name=\\\"rtrp\\\" />\\n            <KeepSequenceSymmetry name=\\\"2state\\\" setting=\\\"true\\\"/>\\n        </TASKOPERATIONS>\\n        <MOVERS>\\n            <SetupForSequenceSymmetryMover name=\\\"almost_linkres\\\" sequence_symmetry_behaviour=\\\"2state\\\" >\\n                <SequenceSymmetry residue_selectors=\\\"chA,chB\\\" />\\n            </SetupForSequenceSymmetryMover>\\n            <SwitchChainOrder name=\\\"delete_Y\\\" chain_order=\\\"1\\\"/>\\n            <SwitchChainOrder name=\\\"delete_X\\\" chain_order=\\\"2\\\"/>\\n            <MutateResidue name=\\\"cys1_X\\\" target=\\\"{cys1_X}\\\" new_res=\\\"CYS\\\" />\\n            <MutateResidue name=\\\"cys2_X\\\" target=\\\"{cys2_X}\\\" new_res=\\\"CYS\\\" />\\n            <MutateResidue name=\\\"cys1_Y\\\" target=\\\"{cys1_Y}\\\" new_res=\\\"CYS\\\" />\\n            <MutateResidue name=\\\"cys2_Y\\\" target=\\\"{cys2_Y}\\\" new_res=\\\"CYS\\\" />\\n            <ForceDisulfides name=\\\"restore_X\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                disulfides=\\\"{cys1_X}:{cys2_X}\\\"\\n                remove_existing=\\\"false\\\"\\n                repack=\\\"true\\\" />\\n            <ForceDisulfides name=\\\"restore_Y\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                disulfides=\\\"{cys1_Y}:{cys2_Y}\\\"\\n                remove_existing=\\\"false\\\"\\n                repack=\\\"true\\\" />\\n            <FastRelax name=\\\"relax\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                repeats=\\\"1\\\" \\n                relaxscript=\\\"MonomerRelax2019\\\"\\n                />\\n            <PackRotamersMover name=\\\"repack\\\" \\n                scorefxn=\\\"sfxn_design\\\"\\n                task_operations=\\\"rtrp\\\"\\n                />\\n        </MOVERS>\\n        <FILTERS>\\n        </FILTERS>\\n        <SIMPLE_METRICS>\\n        </SIMPLE_METRICS>\\n        <APPLY_TO_POSE>\\n        </APPLY_TO_POSE>\\n        <PROTOCOLS>\\n        </PROTOCOLS>\\n    </ROSETTASCRIPTS>\\n    \\\"\\\"\\\".format(\\n        target_resis=target_resis,\\n        cys1_X=cys1_X,\\n        cys2_X=cys2_X,\\n        cys1_Y=cys1_Y,\\n        cys2_Y=cys2_Y,\\n    )\\n    # set taskops by extracting them from the xml_string\\n    xml_obj = XmlObjects.create_from_string(xml_string)\\n    design_task = xml_obj.get_task_operation(\\\"design\\\")\\n    pack_task = xml_obj.get_task_operation(\\\"pack\\\")\\n    lock_task = xml_obj.get_task_operation(\\\"lock\\\")\\n    linkres_task = xml_obj.get_task_operation(\\\"2state\\\")\\n    # set taskops\\n    task_factory = pyrosetta.rosetta.core.pack.task.TaskFactory()\\n    task_factory.push_back(IncludeCurrent())\\n    task_factory.push_back(InitializeFromCommandline())\\n    arochi = LimitAromaChi2Operation()\\n    arochi.include_trp(True)\\n    task_factory.push_back(arochi)\\n    ex1_ex2 = ExtraRotamersGeneric()\\n    ex1_ex2.ex1(True)\\n    ex1_ex2.ex2(True)\\n    task_factory.push_back(ex1_ex2)\\n    task_factory.push_back(design_task)\\n    task_factory.push_back(pack_task)\\n    task_factory.push_back(lock_task)\\n    task_factory.push_back(linkres_task)\\n    # set movemap\\n    mm = pyrosetta.rosetta.core.kinematics.MoveMap()\\n    mm.set_bb(False)\\n    mm.set_chi(True)\\n    mm.set_jump(False)\\n    sfxn_design = xml_obj.get_score_function(\\\"sfxn_design\\\")\\n    fast_design = FastDesign(scorefxn_in=sfxn_design, standard_repeats=1)\\n    fast_design.cartesian(False)\\n    fast_design.set_task_factory(task_factory)\\n    fast_design.set_movemap(mm)\\n    fast_design.minimize_bond_angles(False)\\n    fast_design.minimize_bond_lengths(False)\\n    fast_design.min_type(\\\"lbfgs_armijo_nonmonotone\\\")\\n    fast_design.ramp_down_constraints(False)\\n\\n    name = pose.pdb_info().name()\\n    basename_no_ext = name.split(\\\"/\\\")[-1].replace(\\\".pdb\\\", \\\"\\\", 1)\\n    if out_path[-1] != \\\"/\\\":\\n        out_path += \\\"/\\\"\\n    else:\\n        pass\\n    pose = pose.clone()\\n    almost_linkres = xml_obj.get_mover(\\\"almost_linkres\\\")\\n    almost_linkres.apply(pose)\\n    fast_design.apply(pose)\\n\\n    ref_pose = pose.clone()\\n    # get state X\\n    delete_Y = xml_obj.get_mover(\\\"delete_Y\\\")\\n    delete_Y.apply(pose)\\n    ref_X = pose.clone()\\n    cys1 = xml_obj.get_mover(\\\"cys1_X\\\")\\n    cys2 = xml_obj.get_mover(\\\"cys2_X\\\")\\n    restore = xml_obj.get_mover(\\\"restore_X\\\")\\n    relax = xml_obj.get_mover(\\\"relax\\\")\\n    repack = xml_obj.get_mover(\\\"repack\\\")\\n    cys1.apply(pose)\\n    cys2.apply(pose)\\n    restore.apply(pose)\\n    repack.apply(pose)\\n    # dump X\\n    pose.dump_pdb(os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_X.pdb\\\"))\\n    scores[\\\"X_seq\\\"] = pose.sequence()\\n    # get rmsd\\n    relax_X = ref_X.clone()\\n    relax.apply(relax_X)\\n    # dump X_nocys\\n    relax_X.dump_pdb(\\n        os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_X_nocys.pdb\\\")\\n    )\\n    rmsd_calc = pyrosetta.rosetta.core.simple_metrics.metrics.RMSDMetric()\\n    rmsd_calc.set_rmsd_type(pyrosetta.rosetta.core.scoring.rmsd_atoms(3))\\n    rmsd_calc.set_run_superimpose(True)\\n    rmsd_calc.set_comparison_pose(ref_X)\\n    rmsd = rmsd_calc.calculate(relax_X)\\n    scores[\\\"rmsd_final_X\\\"] = rmsd\\n    # get state Y\\n    pose = ref_pose.clone()\\n    delete_X = xml_obj.get_mover(\\\"delete_X\\\")\\n    delete_X.apply(pose)\\n    ref_Y = pose.clone()\\n    cys1 = xml_obj.get_mover(\\\"cys1_Y\\\")\\n    cys2 = xml_obj.get_mover(\\\"cys2_Y\\\")\\n    restore = xml_obj.get_mover(\\\"restore_Y\\\")\\n    cys1.apply(pose)\\n    cys2.apply(pose)\\n    restore.apply(pose)\\n    repack.apply(pose)\\n    # dump Y\\n    pose.dump_pdb(os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_Y.pdb\\\"))\\n    scores[\\\"Y_seq\\\"] = pose.sequence()\\n    # get rmsd\\n    relax_Y = ref_Y.clone()\\n    relax.apply(relax_Y)\\n    # dump Y_nocys\\n    relax_Y.dump_pdb(\\n        os.path.join(os.getcwd(), out_path + f\\\"{basename_no_ext}_Y_nocys.pdb\\\")\\n    )\\n    rmsd_calc = pyrosetta.rosetta.core.simple_metrics.metrics.RMSDMetric()\\n    rmsd_calc.set_rmsd_type(pyrosetta.rosetta.core.scoring.rmsd_atoms(3))\\n    rmsd_calc.set_run_superimpose(True)\\n    rmsd_calc.set_comparison_pose(ref_Y)\\n    rmsd = rmsd_calc.calculate(relax_Y)\\n    scores[\\\"rmsd_final_Y\\\"] = rmsd\\n    assert relax_X.sequence() == relax_Y.sequence()\\n    scores[\\\"final_sequence\\\"] = relax_X.sequence()\\n    # put in paths to X, Y, X_nocys, Y_nocys\\n    for suffix in [\\\"X\\\", \\\"Y\\\", \\\"X_nocys\\\", \\\"Y_nocys\\\"]:\\n        scores[f\\\"{suffix}_path\\\"] = os.path.join(\\n            os.getcwd(), out_path + f\\\"{basename_no_ext}_{suffix}.pdb\\\"\\n        )\\n    # score relax_X and relax_Y, update scores with those scores\\n    scored_X = score(io.to_packed(relax_X), score_dict=scores)\\n    scored_Y = score(io.to_packed(relax_Y), score_dict=scores)\\n    updates = []\\n    keys = [\\\"HIS_burial\\\", \\\"TRP_burial\\\", \\\"buried_HIS\\\", \\\"near_int_GLY\\\", \\\"near_TRP_aro\\\"]\\n\\n    for chain, score_dict in [\\n        (\\\"X\\\", dict(scored_X.pose.scores)),\\n        (\\\"Y\\\", dict(scored_Y.pose.scores)),\\n    ]:\\n        updates.append(\\n            {\\n                f\\\"{key}_{chain}\\\": value\\n                for key, value in score_dict.items()\\n                if key in keys\\n            }\\n        )\\n    scores.update({**updates[0], **updates[1]})\\n    for key, value in scores.items():\\n        pyrosetta.rosetta.core.pose.setPoseExtraScore(ref_pose, key, value)\\n    ppose = io.to_packed(ref_pose)\\n    return ppose\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from pyrosetta.distributed.packed_pose.core import PackedPose\n",
    "\n",
    "\n",
    "def almost_linkres(packed_pose_in=None, **kwargs) -> PackedPose:\n",
    "    \"\"\"\n",
    "    Load pose. Design with linkres x3: state_X_nocys, state_Y_nocys,\n",
    "    state_XY_nocys. Dump X and Y poses with and without disulfides (4 total).\n",
    "    Return XY without the disulfides.\n",
    "    \"\"\"\n",
    "    import os\n",
    "    import pyrosetta\n",
    "    import pyrosetta.distributed.io as io\n",
    "    from pyrosetta.distributed import cluster\n",
    "    from pyrosetta.rosetta.core.pack.task.operation import (\n",
    "        IncludeCurrent,\n",
    "        InitializeFromCommandline,\n",
    "        ExtraRotamersGeneric,\n",
    "    )\n",
    "    from pyrosetta.rosetta.protocols.denovo_design.movers import FastDesign\n",
    "    from pyrosetta.rosetta.protocols.rosetta_scripts import XmlObjects\n",
    "    from pyrosetta.rosetta.protocols.task_operations import LimitAromaChi2Operation\n",
    "\n",
    "    def parse_scoreinfo(pdb_path) -> dict:\n",
    "        \"\"\"\n",
    "        Hacky unsafe way to load scoreinfo written to pdb by rosetta.\n",
    "        Probably should add additional pdb strings to skip\n",
    "        \"\"\"\n",
    "        with open(pdb_path, \"r\") as f:\n",
    "            scores = {}\n",
    "            pdb_strings = [\"HEADER\", \"EXPDTA\", \"SSBOND\", \"ATOM\", \"TER\", \"REMARK\", \"#\"]\n",
    "            for line in f:\n",
    "                if line == \"\\n\":  # skip empty lines\n",
    "                    pass\n",
    "                # skip pdb specific rows\n",
    "                elif any(pdb_string in line for pdb_string in pdb_strings):\n",
    "                    pass\n",
    "                else:\n",
    "                    key_value = line.rstrip().split()\n",
    "                    # handle pose energies table\n",
    "                    try:\n",
    "                        assert len(key_value) == 2\n",
    "                    except AssertionError:  # pose energies have many values per (residue) key\n",
    "                        continue\n",
    "                    key, value = tuple(key_value)\n",
    "                    if any(c.isalpha() for c in value):  # handle as str below\n",
    "                        pass\n",
    "                    elif \",\" in value:  # handle as str below\n",
    "                        pass\n",
    "                    elif \".\" in value:  # if it has a . it maybe a float\n",
    "                        try:\n",
    "                            value = float(value)\n",
    "                        except ValueError:\n",
    "                            pass\n",
    "                    else:  # if all else fails maybe an int\n",
    "                        try:\n",
    "                            value = int(value)\n",
    "                        except ValueError:\n",
    "                            pass\n",
    "                    scores[key] = value  # if all typecasting failed it is still a str\n",
    "        return scores\n",
    "\n",
    "    def score(packed_pose_in: PackedPose, **kwargs) -> PackedPose:\n",
    "        \"\"\"\n",
    "        Sequence\n",
    "        Compute HIS SASA for X, Y, (compute delta on df)\n",
    "        Compute buried HIS (probably in interface)\n",
    "        Count GLY near interface, (compute glycine count on df)\n",
    "        Compute TRP SASA for X, Y, (compute delta on df)\n",
    "        Count TRP nearby aromatic non-TRP neighbors\n",
    "        Assumes middle is where the switch is split originally, enabling forward and backward SSElement selection approach\n",
    "        @pleung\n",
    "        \"\"\"\n",
    "        import pyrosetta\n",
    "        import pyrosetta.distributed.io as io\n",
    "        from pyrosetta.distributed.tasks.rosetta_scripts import (\n",
    "            SingleoutputRosettaScriptsTask,\n",
    "        )\n",
    "\n",
    "        scores = kwargs[\"score_dict\"]\n",
    "        xml = \"\"\"\n",
    "        <ROSETTASCRIPTS>\n",
    "            <SCOREFXNS>\n",
    "            </SCOREFXNS>\n",
    "            <RESIDUE_SELECTORS>\n",
    "                <ResidueName name=\"GLY\" residue_name3=\"GLY\" />\n",
    "                <ResidueName name=\"HIS\" residue_name3=\"HIS\" />\n",
    "                <ResidueName name=\"TRP\" residue_name3=\"TRP\" />\n",
    "                <ResidueName name=\"aro\" residue_name3=\"TYR,PHE,HIS\" />\n",
    "                <Layer name=\"surface\" select_core=\"false\" select_boundary=\"false\" select_surface=\"true\"\n",
    "                    use_sidechain_neighbors=\"true\"/>\n",
    "                <Layer name=\"boundary\" select_core=\"false\" select_boundary=\"true\" select_surface=\"false\" \n",
    "                    use_sidechain_neighbors=\"true\"/>\n",
    "                <Layer name=\"core\" select_core=\"true\" select_boundary=\"false\" select_surface=\"false\" \n",
    "                    use_sidechain_neighbors=\"true\"/>\n",
    "                <SecondaryStructure name=\"sheet\" overlap=\"0\" minH=\"3\" minE=\"2\" include_terminal_loops=\"false\" \n",
    "                    use_dssp=\"true\" ss=\"E\"/>\n",
    "                <SecondaryStructure name=\"entire_loop\" overlap=\"0\" minH=\"3\" minE=\"2\" include_terminal_loops=\"true\" \n",
    "                    use_dssp=\"true\" ss=\"L\"/>\n",
    "                <SecondaryStructure name=\"entire_helix\" overlap=\"0\" minH=\"3\" minE=\"2\" include_terminal_loops=\"false\"\n",
    "                    use_dssp=\"true\" ss=\"H\"/>\n",
    "                <And name=\"helix_cap\" selectors=\"entire_loop\">\n",
    "                    <PrimarySequenceNeighborhood lower=\"1\" upper=\"0\" selector=\"entire_helix\"/>\n",
    "                </And>\n",
    "                <And name=\"helix_start\" selectors=\"entire_helix\">\n",
    "                    <PrimarySequenceNeighborhood lower=\"0\" upper=\"1\" selector=\"helix_cap\"/>\n",
    "                </And>\n",
    "                <And name=\"helix\" selectors=\"entire_helix\">\n",
    "                    <Not selector=\"helix_start\"/>\n",
    "                </And>\n",
    "                <And name=\"loop\" selectors=\"entire_loop\">\n",
    "                    <Not selector=\"helix_cap\"/>\n",
    "                </And>\n",
    "                <SSElement name=\"near_int\" selection=\"{pre},H,S\" to_selection=\"-{post},H,E\" chain=\"A\" reassign_short_terminal_loop=\"2\" />            \n",
    "                <Neighborhood name=\"near_TRP\" selector=\"TRP\" distance=\"6.0\" />\n",
    "                <And name=\"GLY_near_int\" selectors=\"near_int,HIS\" />\n",
    "                <And name=\"core_HIS\" selectors=\"core,HIS\" />\n",
    "                <And name=\"aro_near_TRP\" selectors=\"aro,near_TRP\" />\n",
    "            </RESIDUE_SELECTORS>\n",
    "            <TASKOPERATIONS>\n",
    "            </TASKOPERATIONS>\n",
    "            <FILTERS>\n",
    "                <BuriedSurfaceArea name=\"HIS_burial\" residue_selector=\"HIS\" confidence=\"0\" />\n",
    "                <BuriedSurfaceArea name=\"TRP_burial\" residue_selector=\"TRP\" confidence=\"0\" />\n",
    "                <ResidueCount name=\"buried_HIS\" residue_selector=\"core_HIS\" />\n",
    "                <ResidueCount name=\"near_int_GLY\" residue_selector=\"GLY_near_int\" />\n",
    "                <ResidueCount name=\"near_TRP_aro\" residue_selector=\"aro_near_TRP\" />\n",
    "            </FILTERS>\n",
    "            <MOVERS>\n",
    "            </MOVERS>\n",
    "            <PROTOCOLS>\n",
    "                <Add filter=\"HIS_burial\" />\n",
    "                <Add filter=\"TRP_burial\" />\n",
    "                <Add filter=\"buried_HIS\" />\n",
    "                <Add filter=\"near_int_GLY\" />\n",
    "                <Add filter=\"near_TRP_aro\" />\n",
    "            </PROTOCOLS>\n",
    "        </ROSETTASCRIPTS>\n",
    "        \"\"\".format(\n",
    "            pre=int(int(scores[\"pre_break_helix\"]) - 1),\n",
    "            post=int(int(scores[\"pre_break_helix\"]) - 1),\n",
    "        )\n",
    "        # relax and score pose\n",
    "        score = SingleoutputRosettaScriptsTask(xml)\n",
    "        scored_ppose = score(packed_pose_in.pose.clone())\n",
    "        update = scored_ppose.pose.scores\n",
    "        pose = io.to_pose(scored_ppose.pose.clone())\n",
    "        # update scores for values that were updated\n",
    "        for key, value in scores.items():\n",
    "            if key in update.keys():\n",
    "                pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, update[key])\n",
    "            else:\n",
    "                pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, value)\n",
    "        final_ppose = io.to_packed(pose)\n",
    "        return final_ppose\n",
    "\n",
    "    if packed_pose_in == None:\n",
    "        pdb_path = kwargs[\"-s\"]\n",
    "    else:\n",
    "        raise RuntimeError(\n",
    "            \"This method requires a pdb path, it cannot accept a PackedPose as input\"\n",
    "        )\n",
    "    scores = parse_scoreinfo(pdb_path)\n",
    "    ppose = io.pose_from_file(pdb_path)\n",
    "    pose = io.to_pose(ppose)\n",
    "    if kwargs[\"out_path\"] == None:\n",
    "        out_path = os.getcwd()\n",
    "    else:\n",
    "        out_path = kwargs[\"out_path\"]\n",
    "        os.makedirs(os.path.join(os.getcwd(), out_path), exist_ok=True)\n",
    "    cys_X = scores[\"disulfide_at_X\"].split(\",\")\n",
    "    cys_Y = scores[\"disulfide_at_Y\"].split(\",\")\n",
    "    unique_cys = set(cys_X + cys_Y)\n",
    "    # check whether disulfides are orthogonal\n",
    "    design_resis = [int(x) for x in unique_cys]\n",
    "    try:\n",
    "        assert len(unique_cys) in [\n",
    "            3,\n",
    "            4,\n",
    "        ]  # if there are less than 3 CYS or more than 4 we really f'd up at some point\n",
    "    except AssertionError:\n",
    "        print(\"There should be at least 3 unique CYS, but no more than 4\")\n",
    "        return None\n",
    "    cys1_X, cys2_X = int(cys_X[0]), int(cys_X[1])\n",
    "    cys1_Y, cys2_Y = int(cys_Y[0]), int(cys_Y[1])\n",
    "    length = scores[\"total_length\"]\n",
    "    resis_matrix = [[x, int(x + length)] for x in design_resis]\n",
    "    target_resis = \",\".join(\n",
    "        [str(x) for row in resis_matrix for x in row]\n",
    "    )  # flatten matrix\n",
    "    xml_string = \"\"\"\n",
    "    <ROSETTASCRIPTS>\n",
    "        <SCOREFXNS>\n",
    "            <ScoreFunction name=\"sfxn_design\" weights=\"beta_nov16\" >\n",
    "                <Set use_hb_env_dep=\"true\" />\n",
    "                <Reweight scoretype=\"approximate_buried_unsat_penalty\" weight=\"17\" />\n",
    "                <Set approximate_buried_unsat_penalty_burial_atomic_depth=\"3.5\" />\n",
    "                <Set approximate_buried_unsat_penalty_hbond_energy_threshold=\"-1.0\" />\n",
    "                <Set approximate_buried_unsat_penalty_natural_corrections1=\"true\" />\n",
    "                <Set approximate_buried_unsat_penalty_hbond_bonus_cross_chain=\"-7\" />\n",
    "                <Set approximate_buried_unsat_penalty_hbond_bonus_ser_to_helix_bb=\"1\"/>                    \n",
    "            </ScoreFunction>\n",
    "        </SCOREFXNS>\n",
    "        <RESIDUE_SELECTORS>\n",
    "            <Chain name=\"chA\" chains=\"A\" />\n",
    "            <Chain name=\"chB\" chains=\"B\" />\n",
    "            <Index name=\"designable\" resnums=\"{target_resis}\" />\n",
    "            <Neighborhood name=\"packable\" selector=\"designable\" distance=\"6.0\" include_focus_in_subset=\"true\" />\n",
    "            <Not name=\"not_designable\" selector=\"designable\" />\n",
    "            <Not name=\"not_packable\" selector=\"packable\" />\n",
    "        </RESIDUE_SELECTORS>\n",
    "        <TASKOPERATIONS>\n",
    "            <OperateOnResidueSubset name=\"design\" selector=\"designable\"> # no CYS; GLY; PRO;\n",
    "                <RestrictAbsentCanonicalAASRLT aas=\"ADEFHIKLMNQRSTVWY\" />\n",
    "            </OperateOnResidueSubset>\n",
    "            <OperateOnResidueSubset name=\"pack\" selector=\"not_designable\">\n",
    "                <RestrictToRepackingRLT/>\n",
    "            </OperateOnResidueSubset>\n",
    "            <OperateOnResidueSubset name=\"lock\" selector=\"not_packable\">\n",
    "                <PreventRepackingRLT/>\n",
    "            </OperateOnResidueSubset>\n",
    "            <RestrictToRepacking name=\"rtrp\" />\n",
    "            <KeepSequenceSymmetry name=\"2state\" setting=\"true\"/>\n",
    "        </TASKOPERATIONS>\n",
    "        <MOVERS>\n",
    "            <SetupForSequenceSymmetryMover name=\"almost_linkres\" sequence_symmetry_behaviour=\"2state\" >\n",
    "                <SequenceSymmetry residue_selectors=\"chA,chB\" />\n",
    "            </SetupForSequenceSymmetryMover>\n",
    "            <SwitchChainOrder name=\"delete_Y\" chain_order=\"1\"/>\n",
    "            <SwitchChainOrder name=\"delete_X\" chain_order=\"2\"/>\n",
    "            <MutateResidue name=\"cys1_X\" target=\"{cys1_X}\" new_res=\"CYS\" />\n",
    "            <MutateResidue name=\"cys2_X\" target=\"{cys2_X}\" new_res=\"CYS\" />\n",
    "            <MutateResidue name=\"cys1_Y\" target=\"{cys1_Y}\" new_res=\"CYS\" />\n",
    "            <MutateResidue name=\"cys2_Y\" target=\"{cys2_Y}\" new_res=\"CYS\" />\n",
    "            <ForceDisulfides name=\"restore_X\" \n",
    "                scorefxn=\"sfxn_design\"\n",
    "                disulfides=\"{cys1_X}:{cys2_X}\"\n",
    "                remove_existing=\"false\"\n",
    "                repack=\"true\" />\n",
    "            <ForceDisulfides name=\"restore_Y\" \n",
    "                scorefxn=\"sfxn_design\"\n",
    "                disulfides=\"{cys1_Y}:{cys2_Y}\"\n",
    "                remove_existing=\"false\"\n",
    "                repack=\"true\" />\n",
    "            <FastRelax name=\"relax\" \n",
    "                scorefxn=\"sfxn_design\"\n",
    "                repeats=\"1\" \n",
    "                relaxscript=\"MonomerRelax2019\"\n",
    "                />\n",
    "            <PackRotamersMover name=\"repack\" \n",
    "                scorefxn=\"sfxn_design\"\n",
    "                task_operations=\"rtrp\"\n",
    "                />\n",
    "        </MOVERS>\n",
    "        <FILTERS>\n",
    "        </FILTERS>\n",
    "        <SIMPLE_METRICS>\n",
    "        </SIMPLE_METRICS>\n",
    "        <APPLY_TO_POSE>\n",
    "        </APPLY_TO_POSE>\n",
    "        <PROTOCOLS>\n",
    "        </PROTOCOLS>\n",
    "    </ROSETTASCRIPTS>\n",
    "    \"\"\".format(\n",
    "        target_resis=target_resis,\n",
    "        cys1_X=cys1_X,\n",
    "        cys2_X=cys2_X,\n",
    "        cys1_Y=cys1_Y,\n",
    "        cys2_Y=cys2_Y,\n",
    "    )\n",
    "    # set taskops by extracting them from the xml_string\n",
    "    xml_obj = XmlObjects.create_from_string(xml_string)\n",
    "    design_task = xml_obj.get_task_operation(\"design\")\n",
    "    pack_task = xml_obj.get_task_operation(\"pack\")\n",
    "    lock_task = xml_obj.get_task_operation(\"lock\")\n",
    "    linkres_task = xml_obj.get_task_operation(\"2state\")\n",
    "    # set taskops\n",
    "    task_factory = pyrosetta.rosetta.core.pack.task.TaskFactory()\n",
    "    task_factory.push_back(IncludeCurrent())\n",
    "    task_factory.push_back(InitializeFromCommandline())\n",
    "    arochi = LimitAromaChi2Operation()\n",
    "    arochi.include_trp(True)\n",
    "    task_factory.push_back(arochi)\n",
    "    ex1_ex2 = ExtraRotamersGeneric()\n",
    "    ex1_ex2.ex1(True)\n",
    "    ex1_ex2.ex2(True)\n",
    "    task_factory.push_back(ex1_ex2)\n",
    "    task_factory.push_back(design_task)\n",
    "    task_factory.push_back(pack_task)\n",
    "    task_factory.push_back(lock_task)\n",
    "    task_factory.push_back(linkres_task)\n",
    "    # set movemap\n",
    "    mm = pyrosetta.rosetta.core.kinematics.MoveMap()\n",
    "    mm.set_bb(False)\n",
    "    mm.set_chi(True)\n",
    "    mm.set_jump(False)\n",
    "    sfxn_design = xml_obj.get_score_function(\"sfxn_design\")\n",
    "    fast_design = FastDesign(scorefxn_in=sfxn_design, standard_repeats=1)\n",
    "    fast_design.cartesian(False)\n",
    "    fast_design.set_task_factory(task_factory)\n",
    "    fast_design.set_movemap(mm)\n",
    "    fast_design.minimize_bond_angles(False)\n",
    "    fast_design.minimize_bond_lengths(False)\n",
    "    fast_design.min_type(\"lbfgs_armijo_nonmonotone\")\n",
    "    fast_design.ramp_down_constraints(False)\n",
    "\n",
    "    name = pose.pdb_info().name()\n",
    "    basename_no_ext = name.split(\"/\")[-1].replace(\".pdb\", \"\", 1)\n",
    "    if out_path[-1] != \"/\":\n",
    "        out_path += \"/\"\n",
    "    else:\n",
    "        pass\n",
    "    pose = pose.clone()\n",
    "    almost_linkres = xml_obj.get_mover(\"almost_linkres\")\n",
    "    almost_linkres.apply(pose)\n",
    "    fast_design.apply(pose)\n",
    "\n",
    "    ref_pose = pose.clone()\n",
    "    # get state X\n",
    "    delete_Y = xml_obj.get_mover(\"delete_Y\")\n",
    "    delete_Y.apply(pose)\n",
    "    ref_X = pose.clone()\n",
    "    cys1 = xml_obj.get_mover(\"cys1_X\")\n",
    "    cys2 = xml_obj.get_mover(\"cys2_X\")\n",
    "    restore = xml_obj.get_mover(\"restore_X\")\n",
    "    relax = xml_obj.get_mover(\"relax\")\n",
    "    repack = xml_obj.get_mover(\"repack\")\n",
    "    cys1.apply(pose)\n",
    "    cys2.apply(pose)\n",
    "    restore.apply(pose)\n",
    "    repack.apply(pose)\n",
    "    # dump X\n",
    "    pose.dump_pdb(os.path.join(os.getcwd(), out_path + f\"{basename_no_ext}_X.pdb\"))\n",
    "    scores[\"X_seq\"] = pose.sequence()\n",
    "    # get rmsd\n",
    "    relax_X = ref_X.clone()\n",
    "    relax.apply(relax_X)\n",
    "    # dump X_nocys\n",
    "    relax_X.dump_pdb(\n",
    "        os.path.join(os.getcwd(), out_path + f\"{basename_no_ext}_X_nocys.pdb\")\n",
    "    )\n",
    "    rmsd_calc = pyrosetta.rosetta.core.simple_metrics.metrics.RMSDMetric()\n",
    "    rmsd_calc.set_rmsd_type(pyrosetta.rosetta.core.scoring.rmsd_atoms(3))\n",
    "    rmsd_calc.set_run_superimpose(True)\n",
    "    rmsd_calc.set_comparison_pose(ref_X)\n",
    "    rmsd = rmsd_calc.calculate(relax_X)\n",
    "    scores[\"rmsd_final_X\"] = rmsd\n",
    "    # get state Y\n",
    "    pose = ref_pose.clone()\n",
    "    delete_X = xml_obj.get_mover(\"delete_X\")\n",
    "    delete_X.apply(pose)\n",
    "    ref_Y = pose.clone()\n",
    "    cys1 = xml_obj.get_mover(\"cys1_Y\")\n",
    "    cys2 = xml_obj.get_mover(\"cys2_Y\")\n",
    "    restore = xml_obj.get_mover(\"restore_Y\")\n",
    "    cys1.apply(pose)\n",
    "    cys2.apply(pose)\n",
    "    restore.apply(pose)\n",
    "    repack.apply(pose)\n",
    "    # dump Y\n",
    "    pose.dump_pdb(os.path.join(os.getcwd(), out_path + f\"{basename_no_ext}_Y.pdb\"))\n",
    "    scores[\"Y_seq\"] = pose.sequence()\n",
    "    # get rmsd\n",
    "    relax_Y = ref_Y.clone()\n",
    "    relax.apply(relax_Y)\n",
    "    # dump Y_nocys\n",
    "    relax_Y.dump_pdb(\n",
    "        os.path.join(os.getcwd(), out_path + f\"{basename_no_ext}_Y_nocys.pdb\")\n",
    "    )\n",
    "    rmsd_calc = pyrosetta.rosetta.core.simple_metrics.metrics.RMSDMetric()\n",
    "    rmsd_calc.set_rmsd_type(pyrosetta.rosetta.core.scoring.rmsd_atoms(3))\n",
    "    rmsd_calc.set_run_superimpose(True)\n",
    "    rmsd_calc.set_comparison_pose(ref_Y)\n",
    "    rmsd = rmsd_calc.calculate(relax_Y)\n",
    "    scores[\"rmsd_final_Y\"] = rmsd\n",
    "    assert relax_X.sequence() == relax_Y.sequence()\n",
    "    scores[\"final_sequence\"] = relax_X.sequence()\n",
    "    # put in paths to X, Y, X_nocys, Y_nocys\n",
    "    for suffix in [\"X\", \"Y\", \"X_nocys\", \"Y_nocys\"]:\n",
    "        scores[f\"{suffix}_path\"] = os.path.join(\n",
    "            os.getcwd(), out_path + f\"{basename_no_ext}_{suffix}.pdb\"\n",
    "        )\n",
    "    # score relax_X and relax_Y, update scores with those scores\n",
    "    scored_X = score(io.to_packed(relax_X), score_dict=scores)\n",
    "    scored_Y = score(io.to_packed(relax_Y), score_dict=scores)\n",
    "    updates = []\n",
    "    keys = [\"HIS_burial\", \"TRP_burial\", \"buried_HIS\", \"near_int_GLY\", \"near_TRP_aro\"]\n",
    "\n",
    "    for chain, score_dict in [\n",
    "        (\"X\", dict(scored_X.pose.scores)),\n",
    "        (\"Y\", dict(scored_Y.pose.scores)),\n",
    "    ]:\n",
    "        updates.append(\n",
    "            {\n",
    "                f\"{key}_{chain}\": value\n",
    "                for key, value in score_dict.items()\n",
    "                if key in keys\n",
    "            }\n",
    "        )\n",
    "    scores.update({**updates[0], **updates[1]})\n",
    "    for key, value in scores.items():\n",
    "        pyrosetta.rosetta.core.pose.setPoseExtraScore(ref_pose, key, value)\n",
    "    ppose = io.to_packed(ref_pose)\n",
    "    return ppose"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Setup dask, set command line options, make tasks and submit to client for cleanup\n",
    "Not using:  \n",
    "`-indexed_structure_store:fragment_store /net/databases/VALL_clustered/connect_chains/ss_grouped_vall_helix_shortLoop.h5`  \n",
    "`-holes:dalphaball\": \"/home/bcov/ppi/tutorial_build/main/source/external/DAlpahBall/DAlphaBall.gcc`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from dask.distributed import Client\n",
    "from dask_jobqueue import SLURMCluster\n",
    "import logging\n",
    "import pwd\n",
    "from pyrosetta.distributed.cluster.core import PyRosettaCluster\n",
    "\n",
    "\n",
    "logging.basicConfig(level=logging.INFO)\n",
    "selected = os.path.join(os.getcwd(), \"05_filter/1trp_good.list\")\n",
    "options = {\n",
    "    \"-out:level\": \"300\",\n",
    "    \"-detect_disulf\": \"false\",\n",
    "    \"-precompute_ig\": \"true\",\n",
    "}\n",
    "\n",
    "\n",
    "def create_tasks(selected, options):\n",
    "    with open(selected, \"r\") as f:\n",
    "        for i, pdb_path in enumerate(f):\n",
    "            tasks = {\"options\": \"-corrections::beta_nov16 true\"}\n",
    "            tasks[\"extra_options\"] = options\n",
    "            tasks[\"-s\"] = pdb_path.rstrip()\n",
    "            tasks[\"out_path\"] = f\"06_states/{int(i / 250)}\"\n",
    "            yield tasks\n",
    "\n",
    "\n",
    "print(\"run the following from your local terminal:\")\n",
    "print(\n",
    "    f\"ssh -L 8000:localhost:8787 {pwd.getpwuid(os.getuid()).pw_name}@{socket.gethostname()}\"\n",
    ")\n",
    "\n",
    "output_path = os.path.join(os.getcwd(), \"06_cleanup\")\n",
    "\n",
    "if __name__ == \"__main__\":\n",
    "    # configure SLURM cluster as a context manager\n",
    "    with SLURMCluster(\n",
    "        cores=1,\n",
    "        processes=1,\n",
    "        job_cpu=1,\n",
    "        memory=\"4GB\",\n",
    "        queue=\"medium\",\n",
    "        walltime=\"23:30:00\",\n",
    "        death_timeout=120,\n",
    "        local_directory=\"$TMPDIR/dask\",\n",
    "        log_directory=\"/mnt/home/pleung/logs/slurm_logs\",\n",
    "        extra=[\"--lifetime\", \"23h\", \"--lifetime-stagger\", \"4m\"],\n",
    "    ) as cluster:\n",
    "        print(cluster.job_script())\n",
    "        # scale between 1-1020 workers,\n",
    "        cluster.adapt(\n",
    "            minimum=1,\n",
    "            maximum=1020,\n",
    "            wait_count=400,  # Number of consecutive times that a worker should be suggested for removal it is removed\n",
    "            interval=\"5s\",  # Time between checks\n",
    "        )\n",
    "        # setup a client to interact with the cluster as a context manager\n",
    "        with Client(cluster) as client:\n",
    "            print(client)\n",
    "            PyRosettaCluster(\n",
    "                tasks=create_tasks(selected, options),\n",
    "                client=client,\n",
    "                scratch_dir=output_path,\n",
    "                output_path=output_path,\n",
    "            ).distribute(protocols=[almost_linkres])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Look at scores\n",
    "Hacky function to load JSON-like data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>HIS_burial_X</th>\n",
       "      <th>HIS_burial_Y</th>\n",
       "      <th>TRP_burial_X</th>\n",
       "      <th>TRP_burial_Y</th>\n",
       "      <th>X_nocys_path</th>\n",
       "      <th>X_path</th>\n",
       "      <th>X_seq</th>\n",
       "      <th>Y_nocys_path</th>\n",
       "      <th>Y_path</th>\n",
       "      <th>Y_seq</th>\n",
       "      <th>...</th>\n",
       "      <th>total_length</th>\n",
       "      <th>total_score</th>\n",
       "      <th>total_score_X</th>\n",
       "      <th>total_score_Y</th>\n",
       "      <th>vbuns_X</th>\n",
       "      <th>vbuns_Y</th>\n",
       "      <th>wnm_all_X</th>\n",
       "      <th>wnm_all_Y</th>\n",
       "      <th>wnm_hlx_X</th>\n",
       "      <th>wnm_hlx_Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_1c3348aaff704bdaa73c6d67fa2db82a.pdb.bz2</th>\n",
       "      <td>191.598892</td>\n",
       "      <td>161.116302</td>\n",
       "      <td>235.250000</td>\n",
       "      <td>221.223358</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TKEEKERIERIEKEVRSPDPENIREAVRKAKELAEENDSEEVKRLL...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TKEEKERIERIEKEVRSPDPENIREAVRKAKELAEENDSCEVKRLL...</td>\n",
       "      <td>...</td>\n",
       "      <td>156.0</td>\n",
       "      <td>-200.714910</td>\n",
       "      <td>-495.980011</td>\n",
       "      <td>-490.601990</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.344967</td>\n",
       "      <td>0.369241</td>\n",
       "      <td>0.121823</td>\n",
       "      <td>0.108301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_3d928f3daa034749a030c7e9ce8f40db.pdb.bz2</th>\n",
       "      <td>191.742218</td>\n",
       "      <td>190.977066</td>\n",
       "      <td>235.254822</td>\n",
       "      <td>219.012299</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TKEEKERIERIEKEVRSPDPENIREAVRKAVELMRENPSEEVKRLV...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TKEEKERIERIEKEVRSPDPENIREAVRKAVELMRENPSCEVKRLV...</td>\n",
       "      <td>...</td>\n",
       "      <td>156.0</td>\n",
       "      <td>15.583651</td>\n",
       "      <td>-518.669983</td>\n",
       "      <td>-505.704010</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.349388</td>\n",
       "      <td>0.696476</td>\n",
       "      <td>0.120602</td>\n",
       "      <td>0.129675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_28be1bd66de54aed83f7639dcd62c32f.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>221.614243</td>\n",
       "      <td>242.966599</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>DSEEEQERIRRILKEARKSGTEESLEQAIEDVAQLAKKSQDPRVLE...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>DSEEEQERIRRILKEARKSGTEESLEQAIEDVAQLAKKSQDPRVLE...</td>\n",
       "      <td>...</td>\n",
       "      <td>160.0</td>\n",
       "      <td>-205.345371</td>\n",
       "      <td>-548.015991</td>\n",
       "      <td>-559.239014</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.433880</td>\n",
       "      <td>0.644063</td>\n",
       "      <td>0.134351</td>\n",
       "      <td>0.108984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_0762f62cd16741048a97cb9036adac58.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>218.944855</td>\n",
       "      <td>230.889313</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TKEEKERIERIEKEVRSPDPENIREAVRKAWELVEENPSEEALRLL...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TKEEKERIERIEKEVRSPDPENIREAVRKAWELVEENPSCEALRLL...</td>\n",
       "      <td>...</td>\n",
       "      <td>156.0</td>\n",
       "      <td>-67.159011</td>\n",
       "      <td>-514.296997</td>\n",
       "      <td>-492.813995</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.353671</td>\n",
       "      <td>0.777966</td>\n",
       "      <td>0.124616</td>\n",
       "      <td>0.113692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_394bedb1240946b9bbaf95c504a5e428.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>191.111252</td>\n",
       "      <td>208.557068</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TIEEIAAELKKTIKEAAEAGADEEAVKDARKLVKEVEKAAKRGVDE...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>TIEEIAAELKKTIKEAAEAGADEEAVKDARKLVKEVEKAAKRGVDE...</td>\n",
       "      <td>...</td>\n",
       "      <td>184.0</td>\n",
       "      <td>14.682345</td>\n",
       "      <td>-566.111023</td>\n",
       "      <td>-566.744995</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.687405</td>\n",
       "      <td>0.673628</td>\n",
       "      <td>0.122836</td>\n",
       "      <td>0.106946</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 145 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    HIS_burial_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    191.598892   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    191.742218   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "\n",
       "                                                    HIS_burial_Y  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    161.116302   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    190.977066   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "\n",
       "                                                    TRP_burial_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    235.250000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    235.254822   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    221.614243   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    218.944855   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    191.111252   \n",
       "\n",
       "                                                    TRP_burial_Y  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    221.223358   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    219.012299   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    242.966599   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    230.889313   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    208.557068   \n",
       "\n",
       "                                                                                         X_nocys_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                               X_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                                X_seq  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TKEEKERIERIEKEVRSPDPENIREAVRKAKELAEENDSEEVKRLL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TKEEKERIERIEKEVRSPDPENIREAVRKAVELMRENPSEEVKRLV...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  DSEEEQERIRRILKEARKSGTEESLEQAIEDVAQLAKKSQDPRVLE...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TKEEKERIERIEKEVRSPDPENIREAVRKAWELVEENPSEEALRLL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TIEEIAAELKKTIKEAAEAGADEEAVKDARKLVKEVEKAAKRGVDE...   \n",
       "\n",
       "                                                                                         Y_nocys_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                               Y_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                                Y_seq  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TKEEKERIERIEKEVRSPDPENIREAVRKAKELAEENDSCEVKRLL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TKEEKERIERIEKEVRSPDPENIREAVRKAVELMRENPSCEVKRLV...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  DSEEEQERIRRILKEARKSGTEESLEQAIEDVAQLAKKSQDPRVLE...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TKEEKERIERIEKEVRSPDPENIREAVRKAWELVEENPSCEALRLL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  TIEEIAAELKKTIKEAAEAGADEEAVKDARKLVKEVEKAAKRGVDE...   \n",
       "\n",
       "                                                    ...  total_length  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         156.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         156.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         160.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         156.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         184.0   \n",
       "\n",
       "                                                    total_score  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  -200.714910   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    15.583651   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  -205.345371   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   -67.159011   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    14.682345   \n",
       "\n",
       "                                                    total_score_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -495.980011   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -518.669983   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -548.015991   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -514.296997   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -566.111023   \n",
       "\n",
       "                                                    total_score_Y  vbuns_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -490.601990      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -505.704010      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -559.239014      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -492.813995      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -566.744995      0.0   \n",
       "\n",
       "                                                    vbuns_Y  wnm_all_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.344967   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.349388   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.433880   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.353671   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.687405   \n",
       "\n",
       "                                                   wnm_all_Y wnm_hlx_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.369241  0.121823   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.696476  0.120602   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.644063  0.134351   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.777966  0.124616   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.673628  0.122836   \n",
       "\n",
       "                                                    wnm_hlx_Y  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.108301  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.129675  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.108984  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.113692  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.106946  \n",
       "\n",
       "[5 rows x 145 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 3;\n",
       "                var nbb_unformatted_code = \"def read_scorefile(scores):\\n    import pandas as pd\\n\\n    scores = pd.read_json(scores, orient=\\\"records\\\", typ=\\\"frame\\\", lines=True)\\n    scores = scores.T\\n    mat = scores.values\\n    n = mat.shape[0]\\n    dicts = list(mat[range(n), range(n)])\\n    index = scores.index\\n    tabulated_scores = pd.DataFrame(dicts, index=index)\\n    return tabulated_scores\\n\\n\\noutput_path = os.path.join(os.getcwd(), \\\"06_cleanup\\\")\\nscores = os.path.join(output_path, \\\"scores.json\\\")\\nscores_df = read_scorefile(scores)\\nscores_df.head()\";\n",
       "                var nbb_formatted_code = \"def read_scorefile(scores):\\n    import pandas as pd\\n\\n    scores = pd.read_json(scores, orient=\\\"records\\\", typ=\\\"frame\\\", lines=True)\\n    scores = scores.T\\n    mat = scores.values\\n    n = mat.shape[0]\\n    dicts = list(mat[range(n), range(n)])\\n    index = scores.index\\n    tabulated_scores = pd.DataFrame(dicts, index=index)\\n    return tabulated_scores\\n\\n\\noutput_path = os.path.join(os.getcwd(), \\\"06_cleanup\\\")\\nscores = os.path.join(output_path, \\\"scores.json\\\")\\nscores_df = read_scorefile(scores)\\nscores_df.head()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def read_scorefile(scores):\n",
    "    import pandas as pd\n",
    "\n",
    "    scores = pd.read_json(scores, orient=\"records\", typ=\"frame\", lines=True)\n",
    "    scores = scores.T\n",
    "    mat = scores.values\n",
    "    n = mat.shape[0]\n",
    "    dicts = list(mat[range(n), range(n)])\n",
    "    index = scores.index\n",
    "    tabulated_scores = pd.DataFrame(dicts, index=index)\n",
    "    return tabulated_scores\n",
    "\n",
    "\n",
    "output_path = os.path.join(os.getcwd(), \"06_cleanup\")\n",
    "scores = os.path.join(output_path, \"scores.json\")\n",
    "scores_df = read_scorefile(scores)\n",
    "scores_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Refilter to remove accidentally introduced TRP\n",
    "Could use `if row[\"final_sequence\"].count(\"AAA\") < 2:` to remove polyALA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "12617it [00:01, 7851.40it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "before refiltering: 12617, after filtering: 12602\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 4;\n",
       "                var nbb_unformatted_code = \"idces = []\\nfor i, row in tqdm(scores_df.iterrows()):\\n    if row[\\\"final_sequence\\\"].count(\\\"W\\\") == 1:\\n        idces.append(i)\\n\\none_trp = scores_df.loc[idces]\\nprint(f\\\"before refiltering: {len(scores_df)}, after filtering: {len(one_trp)}\\\")\";\n",
       "                var nbb_formatted_code = \"idces = []\\nfor i, row in tqdm(scores_df.iterrows()):\\n    if row[\\\"final_sequence\\\"].count(\\\"W\\\") == 1:\\n        idces.append(i)\\n\\none_trp = scores_df.loc[idces]\\nprint(f\\\"before refiltering: {len(scores_df)}, after filtering: {len(one_trp)}\\\")\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "idces = []\n",
    "for i, row in tqdm(scores_df.iterrows()):\n",
    "    if row[\"final_sequence\"].count(\"W\") == 1:\n",
    "        idces.append(i)\n",
    "\n",
    "one_trp = scores_df.loc[idces]\n",
    "print(f\"before refiltering: {len(scores_df)}, after filtering: {len(one_trp)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 5;\n",
       "                var nbb_unformatted_code = \"with open(os.path.join(os.getcwd(), \\\"06_cleanup\\\", \\\"1trp_refiltered.list\\\"), \\\"w\\\") as f:\\n    for i in one_trp.index:\\n        print(i, file=f)\";\n",
       "                var nbb_formatted_code = \"with open(os.path.join(os.getcwd(), \\\"06_cleanup\\\", \\\"1trp_refiltered.list\\\"), \\\"w\\\") as f:\\n    for i in one_trp.index:\\n        print(i, file=f)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with open(os.path.join(os.getcwd(), \"06_cleanup\", \"1trp_refiltered.list\"), \"w\") as f:\n",
    "    for i in one_trp.index:\n",
    "        print(i, file=f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Score final disulfides after dualspace relax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 3;\n",
       "                var nbb_unformatted_code = \"from pyrosetta.distributed.packed_pose.core import PackedPose\\n\\n\\ndef score_dslf(packed_pose_in=None, **kwargs) -> PackedPose:\\n    \\\"\\\"\\\"\\n    Score X and Y disulfides\\n    \\\"\\\"\\\"\\n    import bz2\\n    import pyrosetta\\n    import pyrosetta.distributed.io as io\\n    from pyrosetta.distributed import cluster\\n    from pyrosetta.distributed.tasks.rosetta_scripts import (\\n        SingleoutputRosettaScriptsTask,\\n    )\\n    \\n    \\n    if packed_pose_in == None:\\n        file = kwargs[\\\"-s\\\"]\\n    else:\\n        raise RuntimeError(\\n            \\\"This method requires a pdb path, it cannot accept a PackedPose as input\\\"\\n        )\\n    with open(file, \\\"rb\\\") as f:\\n        ppose = io.pose_from_pdbstring(bz2.decompress(f.read()).decode())\\n    scores = pyrosetta.distributed.cluster.get_scores_dict(file)[\\\"scores\\\"]\\n    \\n    X_ppose = io.pose_from_file(scores[\\\"X_path\\\"])\\n    Y_ppose = io.pose_from_file(scores[\\\"Y_path\\\"])\\n\\n    xml = \\\"\\\"\\\"\\n    <ROSETTASCRIPTS>\\n        <SCOREFXNS>\\n            <ScoreFunction name=\\\"sfxn_design_cart\\\" weights=\\\"beta_nov16_cart\\\" >\\n                <Set use_hb_env_dep=\\\"true\\\" />\\n                <Reweight scoretype=\\\"approximate_buried_unsat_penalty\\\" weight=\\\"17\\\" />\\n                <Set approximate_buried_unsat_penalty_burial_atomic_depth=\\\"3.5\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_energy_threshold=\\\"-1.0\\\" />\\n                <Set approximate_buried_unsat_penalty_natural_corrections1=\\\"true\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_cross_chain=\\\"-7\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_ser_to_helix_bb=\\\"1\\\"/>                    \\n            </ScoreFunction>\\n        </SCOREFXNS>\\n        <RESIDUE_SELECTORS>\\n        </RESIDUE_SELECTORS>\\n        <TASKOPERATIONS>\\n        </TASKOPERATIONS>\\n        <MOVERS>\\n            <FastRelax name=\\\"relax_cart\\\" \\n                scorefxn=\\\"sfxn_design_cart\\\"\\n                repeats=\\\"2\\\" \\n                dualspace=\\\"true\\\"\\n                bondangle=\\\"true\\\"\\n                bondlength=\\\"true\\\"\\n                relaxscript=\\\"MonomerRelax2019\\\"\\n                />\\n        </MOVERS>\\n        <FILTERS>\\n        </FILTERS>\\n        <SIMPLE_METRICS>\\n        </SIMPLE_METRICS>\\n        <APPLY_TO_POSE>\\n        </APPLY_TO_POSE>\\n        <PROTOCOLS>\\n            <Add mover_name=\\\"relax_cart\\\" />\\n        </PROTOCOLS>\\n    </ROSETTASCRIPTS>\\n    \\\"\\\"\\\"\\n    score = SingleoutputRosettaScriptsTask(xml)\\n    scored_X = score(X_ppose.pose.clone())\\n    scored_Y = score(Y_ppose.pose.clone())\\n    scores[\\\"dslf_fa13_X\\\"] = scored_X.pose.scores[\\\"dslf_fa13\\\"]\\n    scores[\\\"dslf_fa13_Y\\\"] = scored_Y.pose.scores[\\\"dslf_fa13\\\"]\\n    pose = io.to_pose(ppose)\\n    for key, value in scores.items():\\n        pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, value)\\n    ppose = io.to_packed(pose)\\n    return ppose\";\n",
       "                var nbb_formatted_code = \"from pyrosetta.distributed.packed_pose.core import PackedPose\\n\\n\\ndef score_dslf(packed_pose_in=None, **kwargs) -> PackedPose:\\n    \\\"\\\"\\\"\\n    Score X and Y disulfides\\n    \\\"\\\"\\\"\\n    import bz2\\n    import pyrosetta\\n    import pyrosetta.distributed.io as io\\n    from pyrosetta.distributed import cluster\\n    from pyrosetta.distributed.tasks.rosetta_scripts import (\\n        SingleoutputRosettaScriptsTask,\\n    )\\n\\n    if packed_pose_in == None:\\n        file = kwargs[\\\"-s\\\"]\\n    else:\\n        raise RuntimeError(\\n            \\\"This method requires a pdb path, it cannot accept a PackedPose as input\\\"\\n        )\\n    with open(file, \\\"rb\\\") as f:\\n        ppose = io.pose_from_pdbstring(bz2.decompress(f.read()).decode())\\n    scores = pyrosetta.distributed.cluster.get_scores_dict(file)[\\\"scores\\\"]\\n\\n    X_ppose = io.pose_from_file(scores[\\\"X_path\\\"])\\n    Y_ppose = io.pose_from_file(scores[\\\"Y_path\\\"])\\n\\n    xml = \\\"\\\"\\\"\\n    <ROSETTASCRIPTS>\\n        <SCOREFXNS>\\n            <ScoreFunction name=\\\"sfxn_design_cart\\\" weights=\\\"beta_nov16_cart\\\" >\\n                <Set use_hb_env_dep=\\\"true\\\" />\\n                <Reweight scoretype=\\\"approximate_buried_unsat_penalty\\\" weight=\\\"17\\\" />\\n                <Set approximate_buried_unsat_penalty_burial_atomic_depth=\\\"3.5\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_energy_threshold=\\\"-1.0\\\" />\\n                <Set approximate_buried_unsat_penalty_natural_corrections1=\\\"true\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_cross_chain=\\\"-7\\\" />\\n                <Set approximate_buried_unsat_penalty_hbond_bonus_ser_to_helix_bb=\\\"1\\\"/>                    \\n            </ScoreFunction>\\n        </SCOREFXNS>\\n        <RESIDUE_SELECTORS>\\n        </RESIDUE_SELECTORS>\\n        <TASKOPERATIONS>\\n        </TASKOPERATIONS>\\n        <MOVERS>\\n            <FastRelax name=\\\"relax_cart\\\" \\n                scorefxn=\\\"sfxn_design_cart\\\"\\n                repeats=\\\"2\\\" \\n                dualspace=\\\"true\\\"\\n                bondangle=\\\"true\\\"\\n                bondlength=\\\"true\\\"\\n                relaxscript=\\\"MonomerRelax2019\\\"\\n                />\\n        </MOVERS>\\n        <FILTERS>\\n        </FILTERS>\\n        <SIMPLE_METRICS>\\n        </SIMPLE_METRICS>\\n        <APPLY_TO_POSE>\\n        </APPLY_TO_POSE>\\n        <PROTOCOLS>\\n            <Add mover_name=\\\"relax_cart\\\" />\\n        </PROTOCOLS>\\n    </ROSETTASCRIPTS>\\n    \\\"\\\"\\\"\\n    score = SingleoutputRosettaScriptsTask(xml)\\n    scored_X = score(X_ppose.pose.clone())\\n    scored_Y = score(Y_ppose.pose.clone())\\n    scores[\\\"dslf_fa13_X\\\"] = scored_X.pose.scores[\\\"dslf_fa13\\\"]\\n    scores[\\\"dslf_fa13_Y\\\"] = scored_Y.pose.scores[\\\"dslf_fa13\\\"]\\n    pose = io.to_pose(ppose)\\n    for key, value in scores.items():\\n        pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, value)\\n    ppose = io.to_packed(pose)\\n    return ppose\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from pyrosetta.distributed.packed_pose.core import PackedPose\n",
    "\n",
    "\n",
    "def score_dslf(packed_pose_in=None, **kwargs) -> PackedPose:\n",
    "    \"\"\"\n",
    "    Score X and Y disulfides\n",
    "    \"\"\"\n",
    "    import bz2\n",
    "    import pyrosetta\n",
    "    import pyrosetta.distributed.io as io\n",
    "    from pyrosetta.distributed import cluster\n",
    "    from pyrosetta.distributed.tasks.rosetta_scripts import (\n",
    "        SingleoutputRosettaScriptsTask,\n",
    "    )\n",
    "    \n",
    "    \n",
    "    if packed_pose_in == None:\n",
    "        file = kwargs[\"-s\"]\n",
    "    else:\n",
    "        raise RuntimeError(\n",
    "            \"This method requires a pdb path, it cannot accept a PackedPose as input\"\n",
    "        )\n",
    "    with open(file, \"rb\") as f:\n",
    "        ppose = io.pose_from_pdbstring(bz2.decompress(f.read()).decode())\n",
    "    scores = pyrosetta.distributed.cluster.get_scores_dict(file)[\"scores\"]\n",
    "    \n",
    "    X_ppose = io.pose_from_file(scores[\"X_path\"])\n",
    "    Y_ppose = io.pose_from_file(scores[\"Y_path\"])\n",
    "\n",
    "    xml = \"\"\"\n",
    "    <ROSETTASCRIPTS>\n",
    "        <SCOREFXNS>\n",
    "            <ScoreFunction name=\"sfxn_design_cart\" weights=\"beta_nov16_cart\" >\n",
    "                <Set use_hb_env_dep=\"true\" />\n",
    "                <Reweight scoretype=\"approximate_buried_unsat_penalty\" weight=\"17\" />\n",
    "                <Set approximate_buried_unsat_penalty_burial_atomic_depth=\"3.5\" />\n",
    "                <Set approximate_buried_unsat_penalty_hbond_energy_threshold=\"-1.0\" />\n",
    "                <Set approximate_buried_unsat_penalty_natural_corrections1=\"true\" />\n",
    "                <Set approximate_buried_unsat_penalty_hbond_bonus_cross_chain=\"-7\" />\n",
    "                <Set approximate_buried_unsat_penalty_hbond_bonus_ser_to_helix_bb=\"1\"/>                    \n",
    "            </ScoreFunction>\n",
    "        </SCOREFXNS>\n",
    "        <RESIDUE_SELECTORS>\n",
    "        </RESIDUE_SELECTORS>\n",
    "        <TASKOPERATIONS>\n",
    "        </TASKOPERATIONS>\n",
    "        <MOVERS>\n",
    "            <FastRelax name=\"relax_cart\" \n",
    "                scorefxn=\"sfxn_design_cart\"\n",
    "                repeats=\"2\" \n",
    "                dualspace=\"true\"\n",
    "                bondangle=\"true\"\n",
    "                bondlength=\"true\"\n",
    "                relaxscript=\"MonomerRelax2019\"\n",
    "                />\n",
    "        </MOVERS>\n",
    "        <FILTERS>\n",
    "        </FILTERS>\n",
    "        <SIMPLE_METRICS>\n",
    "        </SIMPLE_METRICS>\n",
    "        <APPLY_TO_POSE>\n",
    "        </APPLY_TO_POSE>\n",
    "        <PROTOCOLS>\n",
    "            <Add mover_name=\"relax_cart\" />\n",
    "        </PROTOCOLS>\n",
    "    </ROSETTASCRIPTS>\n",
    "    \"\"\"\n",
    "    score = SingleoutputRosettaScriptsTask(xml)\n",
    "    scored_X = score(X_ppose.pose.clone())\n",
    "    scored_Y = score(Y_ppose.pose.clone())\n",
    "    scores[\"dslf_fa13_X\"] = scored_X.pose.scores[\"dslf_fa13\"]\n",
    "    scores[\"dslf_fa13_Y\"] = scored_Y.pose.scores[\"dslf_fa13\"]\n",
    "    pose = io.to_pose(ppose)\n",
    "    for key, value in scores.items():\n",
    "        pyrosetta.rosetta.core.pose.setPoseExtraScore(pose, key, value)\n",
    "    ppose = io.to_packed(pose)\n",
    "    return ppose"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Setup dask, set command line options, make tasks and submit to client for scoring disulfides"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "run the following from your local terminal:\n",
      "ssh -L 8000:localhost:8787 pleung@dig80\n",
      "#!/usr/bin/env bash\n",
      "\n",
      "#SBATCH -J dask-worker\n",
      "#SBATCH -e /mnt/home/pleung/logs/slurm_logs/dask-worker-%J.err\n",
      "#SBATCH -o /mnt/home/pleung/logs/slurm_logs/dask-worker-%J.out\n",
      "#SBATCH -p medium\n",
      "#SBATCH -n 1\n",
      "#SBATCH --cpus-per-task=1\n",
      "#SBATCH --mem=4G\n",
      "#SBATCH -t 23:30:00\n",
      "\n",
      "JOB_ID=${SLURM_JOB_ID%;*}\n",
      "\n",
      "/home/pleung/.conda/envs/phil/bin/python -m distributed.cli.dask_worker tcp://172.16.131.110:7645 --nthreads 1 --memory-limit 3.73GiB --name name --nanny --death-timeout 120 --local-directory $TMPDIR/dask --lifetime 23h --lifetime-stagger 4m\n",
      "\n",
      "<Client: 'tcp://172.16.131.110:7645' processes=0 threads=0, memory=0 B>\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyrosetta.distributed:maybe_init performing pyrosetta initialization: {'options': '-run:constant_seed 1 -multithreading:total_threads 1', 'extra_options': '-mute all', 'set_logging_handler': 'interactive', 'silent': True}\n",
      "INFO:pyrosetta.rosetta:Found rosetta database at: /home/pleung/.conda/envs/phil/lib/python3.8/site-packages/pyrosetta/database; using it....\n",
      "INFO:pyrosetta.rosetta:PyRosetta-4 2021 [Rosetta PyRosetta4.conda.linux.cxx11thread.serialization.CentOS.python38.Release 2021.12+release.ed6a5560506cfd327d4a6a3e2c9b0c9f6f4a6535 2021-03-26T16:09:25] retrieved from: http://www.pyrosetta.org\n",
      "(C) Copyright Rosetta Commons Member Institutions. Created in JHU by Sergey Lyskov and PyRosetta Team.\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 4;\n",
       "                var nbb_unformatted_code = \"from dask.distributed import Client\\nfrom dask_jobqueue import SLURMCluster\\nimport logging\\nimport pwd\\nfrom pyrosetta.distributed.cluster.core import PyRosettaCluster\\n\\n\\nlogging.basicConfig(level=logging.INFO)\\nselected = os.path.join(os.getcwd(), \\\"06_cleanup/1trp_refiltered.list\\\")\\noptions = {\\n    \\\"-out:level\\\": \\\"300\\\",\\n}\\n\\n\\ndef create_tasks(selected, options):\\n    with open(selected, \\\"r\\\") as f:\\n        for i, pdb_path in enumerate(f):\\n            tasks = {\\\"options\\\": \\\"-corrections::beta_nov16 true\\\"}\\n            tasks[\\\"extra_options\\\"] = options\\n            tasks[\\\"-s\\\"] = pdb_path.rstrip()\\n            yield tasks\\n\\n\\nprint(\\\"run the following from your local terminal:\\\")\\nprint(\\n    f\\\"ssh -L 8000:localhost:8787 {pwd.getpwuid(os.getuid()).pw_name}@{socket.gethostname()}\\\"\\n)\\n\\noutput_path = os.path.join(os.getcwd(), \\\"06_score_dslf\\\")\\n\\nif __name__ == \\\"__main__\\\":\\n    # configure SLURM cluster as a context manager\\n    with SLURMCluster(\\n        cores=1,\\n        processes=1,\\n        job_cpu=1,\\n        memory=\\\"4GB\\\",\\n        queue=\\\"medium\\\",\\n        walltime=\\\"23:30:00\\\",\\n        death_timeout=120,\\n        local_directory=\\\"$TMPDIR/dask\\\",\\n        log_directory=\\\"/mnt/home/pleung/logs/slurm_logs\\\",\\n        extra=[\\\"--lifetime\\\", \\\"23h\\\", \\\"--lifetime-stagger\\\", \\\"4m\\\"],\\n    ) as cluster:\\n        print(cluster.job_script())\\n        # scale between 1-1020 workers,\\n        cluster.adapt(\\n            minimum=1,\\n            maximum=1020,\\n            wait_count=400,  # Number of consecutive times that a worker should be suggested for removal it is removed\\n            interval=\\\"5s\\\",  # Time between checks\\n        )\\n        # setup a client to interact with the cluster as a context manager\\n        with Client(cluster) as client:\\n            print(client)\\n            PyRosettaCluster(\\n                tasks=create_tasks(selected, options),\\n                client=client,\\n                scratch_dir=output_path,\\n                output_path=output_path,\\n            ).distribute(protocols=[score_dslf])\";\n",
       "                var nbb_formatted_code = \"from dask.distributed import Client\\nfrom dask_jobqueue import SLURMCluster\\nimport logging\\nimport pwd\\nfrom pyrosetta.distributed.cluster.core import PyRosettaCluster\\n\\n\\nlogging.basicConfig(level=logging.INFO)\\nselected = os.path.join(os.getcwd(), \\\"06_cleanup/1trp_refiltered.list\\\")\\noptions = {\\n    \\\"-out:level\\\": \\\"300\\\",\\n}\\n\\n\\ndef create_tasks(selected, options):\\n    with open(selected, \\\"r\\\") as f:\\n        for i, pdb_path in enumerate(f):\\n            tasks = {\\\"options\\\": \\\"-corrections::beta_nov16 true\\\"}\\n            tasks[\\\"extra_options\\\"] = options\\n            tasks[\\\"-s\\\"] = pdb_path.rstrip()\\n            yield tasks\\n\\n\\nprint(\\\"run the following from your local terminal:\\\")\\nprint(\\n    f\\\"ssh -L 8000:localhost:8787 {pwd.getpwuid(os.getuid()).pw_name}@{socket.gethostname()}\\\"\\n)\\n\\noutput_path = os.path.join(os.getcwd(), \\\"06_score_dslf\\\")\\n\\nif __name__ == \\\"__main__\\\":\\n    # configure SLURM cluster as a context manager\\n    with SLURMCluster(\\n        cores=1,\\n        processes=1,\\n        job_cpu=1,\\n        memory=\\\"4GB\\\",\\n        queue=\\\"medium\\\",\\n        walltime=\\\"23:30:00\\\",\\n        death_timeout=120,\\n        local_directory=\\\"$TMPDIR/dask\\\",\\n        log_directory=\\\"/mnt/home/pleung/logs/slurm_logs\\\",\\n        extra=[\\\"--lifetime\\\", \\\"23h\\\", \\\"--lifetime-stagger\\\", \\\"4m\\\"],\\n    ) as cluster:\\n        print(cluster.job_script())\\n        # scale between 1-1020 workers,\\n        cluster.adapt(\\n            minimum=1,\\n            maximum=1020,\\n            wait_count=400,  # Number of consecutive times that a worker should be suggested for removal it is removed\\n            interval=\\\"5s\\\",  # Time between checks\\n        )\\n        # setup a client to interact with the cluster as a context manager\\n        with Client(cluster) as client:\\n            print(client)\\n            PyRosettaCluster(\\n                tasks=create_tasks(selected, options),\\n                client=client,\\n                scratch_dir=output_path,\\n                output_path=output_path,\\n            ).distribute(protocols=[score_dslf])\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from dask.distributed import Client\n",
    "from dask_jobqueue import SLURMCluster\n",
    "import logging\n",
    "import pwd\n",
    "from pyrosetta.distributed.cluster.core import PyRosettaCluster\n",
    "\n",
    "\n",
    "logging.basicConfig(level=logging.INFO)\n",
    "selected = os.path.join(os.getcwd(), \"06_cleanup/1trp_refiltered.list\")\n",
    "options = {\n",
    "    \"-out:level\": \"300\",\n",
    "}\n",
    "\n",
    "\n",
    "def create_tasks(selected, options):\n",
    "    with open(selected, \"r\") as f:\n",
    "        for i, pdb_path in enumerate(f):\n",
    "            tasks = {\"options\": \"-corrections::beta_nov16 true\"}\n",
    "            tasks[\"extra_options\"] = options\n",
    "            tasks[\"-s\"] = pdb_path.rstrip()\n",
    "            yield tasks\n",
    "\n",
    "\n",
    "print(\"run the following from your local terminal:\")\n",
    "print(\n",
    "    f\"ssh -L 8000:localhost:8787 {pwd.getpwuid(os.getuid()).pw_name}@{socket.gethostname()}\"\n",
    ")\n",
    "\n",
    "output_path = os.path.join(os.getcwd(), \"06_score_dslf\")\n",
    "\n",
    "if __name__ == \"__main__\":\n",
    "    # configure SLURM cluster as a context manager\n",
    "    with SLURMCluster(\n",
    "        cores=1,\n",
    "        processes=1,\n",
    "        job_cpu=1,\n",
    "        memory=\"4GB\",\n",
    "        queue=\"medium\",\n",
    "        walltime=\"23:30:00\",\n",
    "        death_timeout=120,\n",
    "        local_directory=\"$TMPDIR/dask\",\n",
    "        log_directory=\"/mnt/home/pleung/logs/slurm_logs\",\n",
    "        extra=[\"--lifetime\", \"23h\", \"--lifetime-stagger\", \"4m\"],\n",
    "    ) as cluster:\n",
    "        print(cluster.job_script())\n",
    "        # scale between 1-1020 workers,\n",
    "        cluster.adapt(\n",
    "            minimum=1,\n",
    "            maximum=1020,\n",
    "            wait_count=400,  # Number of consecutive times that a worker should be suggested for removal it is removed\n",
    "            interval=\"5s\",  # Time between checks\n",
    "        )\n",
    "        # setup a client to interact with the cluster as a context manager\n",
    "        with Client(cluster) as client:\n",
    "            print(client)\n",
    "            PyRosettaCluster(\n",
    "                tasks=create_tasks(selected, options),\n",
    "                client=client,\n",
    "                scratch_dir=output_path,\n",
    "                output_path=output_path,\n",
    "            ).distribute(protocols=[score_dslf])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reload scores\n",
    "Now with updated disulfide scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>HIS_burial_X</th>\n",
       "      <th>HIS_burial_Y</th>\n",
       "      <th>TRP_burial_X</th>\n",
       "      <th>TRP_burial_Y</th>\n",
       "      <th>X_nocys_path</th>\n",
       "      <th>X_path</th>\n",
       "      <th>X_seq</th>\n",
       "      <th>Y_nocys_path</th>\n",
       "      <th>Y_path</th>\n",
       "      <th>Y_seq</th>\n",
       "      <th>...</th>\n",
       "      <th>total_length</th>\n",
       "      <th>total_score</th>\n",
       "      <th>total_score_X</th>\n",
       "      <th>total_score_Y</th>\n",
       "      <th>vbuns_X</th>\n",
       "      <th>vbuns_Y</th>\n",
       "      <th>wnm_all_X</th>\n",
       "      <th>wnm_all_Y</th>\n",
       "      <th>wnm_hlx_X</th>\n",
       "      <th>wnm_hlx_Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_score_dslf/decoys/0000/2021.07.13.11.08.52.158606_342d143c7be241c5aab45231a6ec2df9.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>195.973801</td>\n",
       "      <td>216.967392</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEVLKVVY...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEVLKVVY...</td>\n",
       "      <td>...</td>\n",
       "      <td>152.0</td>\n",
       "      <td>28.766157</td>\n",
       "      <td>-558.328003</td>\n",
       "      <td>-512.533020</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.406159</td>\n",
       "      <td>0.745775</td>\n",
       "      <td>0.102771</td>\n",
       "      <td>0.092355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_score_dslf/decoys/0000/2021.07.13.11.08.52.158606_448d5681fd3f4da5b32b7da261776daa.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>198.123276</td>\n",
       "      <td>247.703842</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDSDLAKEAQKLADEARKSDDSEALKVVL...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDSDLAKEAQKLADEARKSDDSEALKVVL...</td>\n",
       "      <td>...</td>\n",
       "      <td>152.0</td>\n",
       "      <td>100.806572</td>\n",
       "      <td>-532.127991</td>\n",
       "      <td>-515.400024</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.433829</td>\n",
       "      <td>0.672927</td>\n",
       "      <td>0.137396</td>\n",
       "      <td>0.094759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_score_dslf/decoys/0000/2021.07.13.11.08.52.158606_4e53b7dd92574fd5b74f991789950e64.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>205.448227</td>\n",
       "      <td>174.116074</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSKVLKVVY...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSKVLKVVY...</td>\n",
       "      <td>...</td>\n",
       "      <td>152.0</td>\n",
       "      <td>18.827785</td>\n",
       "      <td>-554.708008</td>\n",
       "      <td>-529.700012</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.426545</td>\n",
       "      <td>0.712637</td>\n",
       "      <td>0.104716</td>\n",
       "      <td>0.103542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_score_dslf/decoys/0000/2021.07.13.11.08.52.158606_e43cabd8f029440d9f7b25c37629f90b.pdb.bz2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>170.717422</td>\n",
       "      <td>231.323502</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTDLAKEAQKLADEARKSDDSRQLKVVL...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTDLAKEAQKLADEARKSDDSRQLKVVL...</td>\n",
       "      <td>...</td>\n",
       "      <td>152.0</td>\n",
       "      <td>64.353439</td>\n",
       "      <td>-542.125977</td>\n",
       "      <td>-500.825012</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.421666</td>\n",
       "      <td>0.790040</td>\n",
       "      <td>0.108086</td>\n",
       "      <td>0.101851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_score_dslf/decoys/0000/2021.07.13.11.08.52.158606_dd71e36d208e430e9e86ed401ff13b3a.pdb.bz2</th>\n",
       "      <td>215.227417</td>\n",
       "      <td>199.043869</td>\n",
       "      <td>175.776077</td>\n",
       "      <td>181.016541</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEQLKVVY...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>/mnt/home/pleung/projects/bistable_bundle/r4/h...</td>\n",
       "      <td>SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEQLKVVY...</td>\n",
       "      <td>...</td>\n",
       "      <td>152.0</td>\n",
       "      <td>27.962423</td>\n",
       "      <td>-561.476990</td>\n",
       "      <td>-512.924011</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.425074</td>\n",
       "      <td>0.637930</td>\n",
       "      <td>0.118128</td>\n",
       "      <td>0.101921</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 145 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    HIS_burial_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    215.227417   \n",
       "\n",
       "                                                    HIS_burial_Y  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.000000   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    199.043869   \n",
       "\n",
       "                                                    TRP_burial_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    195.973801   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    198.123276   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    205.448227   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    170.717422   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    175.776077   \n",
       "\n",
       "                                                    TRP_burial_Y  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    216.967392   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    247.703842   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    174.116074   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    231.323502   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    181.016541   \n",
       "\n",
       "                                                                                         X_nocys_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                               X_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                                X_seq  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEVLKVVY...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDSDLAKEAQKLADEARKSDDSEALKVVL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSKVLKVVY...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTDLAKEAQKLADEARKSDDSRQLKVVL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEQLKVVY...   \n",
       "\n",
       "                                                                                         Y_nocys_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                               Y_path  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  /mnt/home/pleung/projects/bistable_bundle/r4/h...   \n",
       "\n",
       "                                                                                                Y_seq  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEVLKVVY...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDSDLAKEAQKLADEARKSDDSEALKVVL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSKVLKVVY...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTDLAKEAQKLADEARKSDDSRQLKVVL...   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  SEKEKVEELAQRIREQLPDTELAREAQELADEARKSDDSEQLKVVY...   \n",
       "\n",
       "                                                    ...  total_length  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         152.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         152.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         152.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         152.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  ...         152.0   \n",
       "\n",
       "                                                    total_score  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    28.766157   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   100.806572   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    18.827785   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    64.353439   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    27.962423   \n",
       "\n",
       "                                                    total_score_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -558.328003   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -532.127991   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -554.708008   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -542.125977   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -561.476990   \n",
       "\n",
       "                                                    total_score_Y  vbuns_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -512.533020      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -515.400024      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -529.700012      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -500.825012      0.0   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...    -512.924011      0.0   \n",
       "\n",
       "                                                    vbuns_Y  wnm_all_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.406159   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.433829   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.426545   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.421666   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...      0.0   0.425074   \n",
       "\n",
       "                                                   wnm_all_Y wnm_hlx_X  \\\n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.745775  0.102771   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.672927  0.137396   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.712637  0.104716   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.790040  0.108086   \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...  0.637930  0.118128   \n",
       "\n",
       "                                                    wnm_hlx_Y  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.092355  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.094759  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.103542  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.101851  \n",
       "/mnt/home/pleung/projects/bistable_bundle/r4/hi...   0.101921  \n",
       "\n",
       "[5 rows x 145 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 5;\n",
       "                var nbb_unformatted_code = \"def read_scorefile(scores):\\n    import pandas as pd\\n\\n    scores = pd.read_json(scores, orient=\\\"records\\\", typ=\\\"frame\\\", lines=True)\\n    scores = scores.T\\n    mat = scores.values\\n    n = mat.shape[0]\\n    dicts = list(mat[range(n), range(n)])\\n    index = scores.index\\n    tabulated_scores = pd.DataFrame(dicts, index=index)\\n    return tabulated_scores\\n\\n\\noutput_path = os.path.join(os.getcwd(), \\\"06_score_dslf\\\")\\nscores = os.path.join(output_path, \\\"scores.json\\\")\\nscores_df = read_scorefile(scores)\\nscores_df.head()\";\n",
       "                var nbb_formatted_code = \"def read_scorefile(scores):\\n    import pandas as pd\\n\\n    scores = pd.read_json(scores, orient=\\\"records\\\", typ=\\\"frame\\\", lines=True)\\n    scores = scores.T\\n    mat = scores.values\\n    n = mat.shape[0]\\n    dicts = list(mat[range(n), range(n)])\\n    index = scores.index\\n    tabulated_scores = pd.DataFrame(dicts, index=index)\\n    return tabulated_scores\\n\\n\\noutput_path = os.path.join(os.getcwd(), \\\"06_score_dslf\\\")\\nscores = os.path.join(output_path, \\\"scores.json\\\")\\nscores_df = read_scorefile(scores)\\nscores_df.head()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def read_scorefile(scores):\n",
    "    import pandas as pd\n",
    "\n",
    "    scores = pd.read_json(scores, orient=\"records\", typ=\"frame\", lines=True)\n",
    "    scores = scores.T\n",
    "    mat = scores.values\n",
    "    n = mat.shape[0]\n",
    "    dicts = list(mat[range(n), range(n)])\n",
    "    index = scores.index\n",
    "    tabulated_scores = pd.DataFrame(dicts, index=index)\n",
    "    return tabulated_scores\n",
    "\n",
    "\n",
    "output_path = os.path.join(os.getcwd(), \"06_score_dslf\")\n",
    "scores = os.path.join(output_path, \"scores.json\")\n",
    "scores_df = read_scorefile(scores)\n",
    "scores_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analyze sequence and residue metrics\n",
    "\n",
    "GLY in middle loop GLY in interface, GLY positions, DSSP (compute glycine count on df)\n",
    "\n",
    "TRP SASA X, Y, (compute delta on df)\n",
    "TRP nearby aromatic non-TRP neighbors\n",
    "\n",
    "HIS in interface SASA X, Y, (compute delta on df)  \n",
    "HIS in interface\n",
    "\n",
    "\n",
    "(compute sequence recovery of parent in df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 7;\n",
       "                var nbb_unformatted_code = \"sns.set(\\n    context=\\\"talk\\\",\\n    font_scale=1.5,  # make the font larger; default is pretty small\\n    style=\\\"ticks\\\",  # make the background white with black lines\\n    palette=\\\"colorblind\\\",  # a color palette that is colorblind friendly!\\n)\\n\\nfig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"rmsd_final_X\\\", y=\\\"rmsd_final_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_rmsd_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_formatted_code = \"sns.set(\\n    context=\\\"talk\\\",\\n    font_scale=1.5,  # make the font larger; default is pretty small\\n    style=\\\"ticks\\\",  # make the background white with black lines\\n    palette=\\\"colorblind\\\",  # a color palette that is colorblind friendly!\\n)\\n\\nfig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"rmsd_final_X\\\", y=\\\"rmsd_final_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_rmsd_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(\n",
    "    context=\"talk\",\n",
    "    font_scale=1.5,  # make the font larger; default is pretty small\n",
    "    style=\"ticks\",  # make the background white with black lines\n",
    "    palette=\"colorblind\",  # a color palette that is colorblind friendly!\n",
    ")\n",
    "\n",
    "fig = plt.figure(figsize=(20, 20), tight_layout=True)\n",
    "sns.histplot(x=\"rmsd_final_X\", y=\"rmsd_final_Y\", data=scores_df, bins=20)\n",
    "fig.savefig(\"figs/06_rmsd_final.png\")\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 13;\n",
       "                var nbb_unformatted_code = \"fig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"TRP_burial_X\\\", y=\\\"TRP_burial_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_TRP_burial_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_formatted_code = \"fig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"TRP_burial_X\\\", y=\\\"TRP_burial_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_TRP_burial_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20, 20), tight_layout=True)\n",
    "sns.histplot(x=\"TRP_burial_X\", y=\"TRP_burial_Y\", data=scores_df, bins=20)\n",
    "fig.savefig(\"figs/06_TRP_burial_final.png\")\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 12;\n",
       "                var nbb_unformatted_code = \"fig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"HIS_burial_X\\\", y=\\\"HIS_burial_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_HIS_burial_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_formatted_code = \"fig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"HIS_burial_X\\\", y=\\\"HIS_burial_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_HIS_burial_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20, 20), tight_layout=True)\n",
    "sns.histplot(x=\"HIS_burial_X\", y=\"HIS_burial_Y\", data=scores_df, bins=20)\n",
    "fig.savefig(\"figs/06_HIS_burial_final.png\")\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 14;\n",
       "                var nbb_unformatted_code = \"fig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"dslf_fa13_X\\\", y=\\\"dslf_fa13_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_dslf_fa13_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_formatted_code = \"fig = plt.figure(figsize=(20, 20), tight_layout=True)\\nsns.histplot(x=\\\"dslf_fa13_X\\\", y=\\\"dslf_fa13_Y\\\", data=scores_df, bins=20)\\nfig.savefig(\\\"figs/06_dslf_fa13_final.png\\\")\\nplt.close()\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20, 20), tight_layout=True)\n",
    "sns.histplot(x=\"dslf_fa13_X\", y=\"dslf_fa13_Y\", data=scores_df, bins=20)\n",
    "fig.savefig(\"figs/06_dslf_fa13_final.png\")\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2304x2304 with 14 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 16;\n",
       "                var nbb_unformatted_code = \"def rho(x, y, ax=None, **kwargs):\\n    \\\"\\\"\\\"Plot the correlation coefficient in the top left hand corner of a plot.\\n    https://stackoverflow.com/questions/50832204/show-correlation-values-in-pairplot-using-seaborn-in-python/50835066\\n    \\\"\\\"\\\"\\n    import scipy\\n\\n    r, _ = scipy.stats.pearsonr(x, y)\\n    ax = ax or plt.gca()\\n    # Unicode for lowercase rho (\\u03c1)\\n    rho = \\\"\\\\u03C1\\\"\\n    ax.annotate(f\\\"{rho} = {r:.2f}\\\", xy=(0.1, 0.9), xycoords=ax.transAxes)\\n    \\ndslf_subset = scores_df[\\n    [\\n        \\\"dslf_fa13_X\\\",\\n        \\\"dslf_fa13_Y\\\",\\n        \\\"dslf_fa13_cart_X\\\",\\n        \\\"dslf_fa13_cart_Y\\\",\\n        \\\"parent\\\",\\n    ]\\n]\\n\\n# ax = sns.pairplot(\\n#     data=dslf_subset.sample(frac=0.1), hue=\\\"parent\\\", corner=True, height=8\\n# )\\n# plt.suptitle(\\\"Correlation of dslf, split by parent\\\")\\n# sns.despine()\\n# plt.savefig(\\\"figs/06_correlations_dlsf_split_by_parent.png\\\")\\n\\n# plt.close()\\n\\nax = sns.pairplot(data=dslf_subset.sample(frac=0.1), corner=True, height=8)\\nax.map_lower(rho)\\nplt.suptitle(\\\"Correlation of dslf, with pearson R\\\")\\nsns.despine()\\nplt.savefig(\\\"figs/06_correlations_dlsf_pearson.png\\\")\";\n",
       "                var nbb_formatted_code = \"def rho(x, y, ax=None, **kwargs):\\n    \\\"\\\"\\\"Plot the correlation coefficient in the top left hand corner of a plot.\\n    https://stackoverflow.com/questions/50832204/show-correlation-values-in-pairplot-using-seaborn-in-python/50835066\\n    \\\"\\\"\\\"\\n    import scipy\\n\\n    r, _ = scipy.stats.pearsonr(x, y)\\n    ax = ax or plt.gca()\\n    # Unicode for lowercase rho (\\u03c1)\\n    rho = \\\"\\\\u03C1\\\"\\n    ax.annotate(f\\\"{rho} = {r:.2f}\\\", xy=(0.1, 0.9), xycoords=ax.transAxes)\\n\\n\\ndslf_subset = scores_df[\\n    [\\n        \\\"dslf_fa13_X\\\",\\n        \\\"dslf_fa13_Y\\\",\\n        \\\"dslf_fa13_cart_X\\\",\\n        \\\"dslf_fa13_cart_Y\\\",\\n        \\\"parent\\\",\\n    ]\\n]\\n\\n# ax = sns.pairplot(\\n#     data=dslf_subset.sample(frac=0.1), hue=\\\"parent\\\", corner=True, height=8\\n# )\\n# plt.suptitle(\\\"Correlation of dslf, split by parent\\\")\\n# sns.despine()\\n# plt.savefig(\\\"figs/06_correlations_dlsf_split_by_parent.png\\\")\\n\\n# plt.close()\\n\\nax = sns.pairplot(data=dslf_subset.sample(frac=0.1), corner=True, height=8)\\nax.map_lower(rho)\\nplt.suptitle(\\\"Correlation of dslf, with pearson R\\\")\\nsns.despine()\\nplt.savefig(\\\"figs/06_correlations_dlsf_pearson.png\\\")\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def rho(x, y, ax=None, **kwargs):\n",
    "    \"\"\"Plot the correlation coefficient in the top left hand corner of a plot.\n",
    "    https://stackoverflow.com/questions/50832204/show-correlation-values-in-pairplot-using-seaborn-in-python/50835066\n",
    "    \"\"\"\n",
    "    import scipy\n",
    "\n",
    "    r, _ = scipy.stats.pearsonr(x, y)\n",
    "    ax = ax or plt.gca()\n",
    "    # Unicode for lowercase rho (ρ)\n",
    "    rho = \"\\u03C1\"\n",
    "    ax.annotate(f\"{rho} = {r:.2f}\", xy=(0.1, 0.9), xycoords=ax.transAxes)\n",
    "    \n",
    "dslf_subset = scores_df[\n",
    "    [\n",
    "        \"dslf_fa13_X\",\n",
    "        \"dslf_fa13_Y\",\n",
    "        \"dslf_fa13_cart_X\",\n",
    "        \"dslf_fa13_cart_Y\",\n",
    "        \"parent\",\n",
    "    ]\n",
    "]\n",
    "\n",
    "# ax = sns.pairplot(\n",
    "#     data=dslf_subset.sample(frac=0.1), hue=\"parent\", corner=True, height=8\n",
    "# )\n",
    "# plt.suptitle(\"Correlation of dslf, split by parent\")\n",
    "# sns.despine()\n",
    "# plt.savefig(\"figs/06_correlations_dlsf_split_by_parent.png\")\n",
    "\n",
    "# plt.close()\n",
    "\n",
    "ax = sns.pairplot(data=dslf_subset.sample(frac=0.1), corner=True, height=8)\n",
    "ax.map_lower(rho)\n",
    "plt.suptitle(\"Correlation of dslf, with pearson R\")\n",
    "sns.despine()\n",
    "plt.savefig(\"figs/06_correlations_dlsf_pearson.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 17;\n",
       "                var nbb_unformatted_code = \"scores_df[\\\"state\\\"] = (\\n    scores_df[\\\"parent\\\"].astype(str)\\n    + \\\"_p_\\\"\\n    + scores_df[\\\"pivot_helix\\\"].astype(int).astype(str)\\n    + \\\"_s_\\\"\\n    + scores_df[\\\"shift\\\"].astype(int).astype(str)\\n)\";\n",
       "                var nbb_formatted_code = \"scores_df[\\\"state\\\"] = (\\n    scores_df[\\\"parent\\\"].astype(str)\\n    + \\\"_p_\\\"\\n    + scores_df[\\\"pivot_helix\\\"].astype(int).astype(str)\\n    + \\\"_s_\\\"\\n    + scores_df[\\\"shift\\\"].astype(int).astype(str)\\n)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "scores_df[\"state\"] = (\n",
    "    scores_df[\"parent\"].astype(str)\n",
    "    + \"_p_\"\n",
    "    + scores_df[\"pivot_helix\"].astype(int).astype(str)\n",
    "    + \"_s_\"\n",
    "    + scores_df[\"shift\"].astype(int).astype(str)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get the best disulfides"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "29\n",
      "366\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'DHR03_nocys',\n",
       " 'DHR15_nocys',\n",
       " 'DHR18_nocys',\n",
       " 'DHR39',\n",
       " 'DHR46',\n",
       " 'DHR57',\n",
       " 'DHR76',\n",
       " 'DHR79',\n",
       " 'DHR82',\n",
       " 'TH_DHR_T7',\n",
       " 'hDHR14_5CWH',\n",
       " 'hDHR14_5H7C',\n",
       " 'hDHR53_5CWK',\n",
       " 'hDHR64_5CWM',\n",
       " 'hDHR71_5CWN',\n",
       " 'hDHR79_5CWP',\n",
       " 'hDHR9_HFuse',\n",
       " 'hTH_DHR_8_NSR',\n",
       " 'hTH_DHR_DN_T6'}"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 37;\n",
       "                var nbb_unformatted_code = \"best = scores_df[scores_df[\\\"dslf_fa13_X\\\"] < -1]\\nbest = best[best[\\\"dslf_fa13_Y\\\"] < -1]\\nbest = best[best[\\\"rmsd_final_X\\\"] < 0.5]\\nbest = best[best[\\\"rmsd_final_Y\\\"] < 0.5]\\n\\nidces = list(best.index)\\n\\nprint(len(set(best.state.values)))\\nprint(len(best))\\nset(scores_df.parent.values) - set(best.parent.values)\";\n",
       "                var nbb_formatted_code = \"best = scores_df[scores_df[\\\"dslf_fa13_X\\\"] < -1]\\nbest = best[best[\\\"dslf_fa13_Y\\\"] < -1]\\nbest = best[best[\\\"rmsd_final_X\\\"] < 0.5]\\nbest = best[best[\\\"rmsd_final_Y\\\"] < 0.5]\\n\\nidces = list(best.index)\\n\\nprint(len(set(best.state.values)))\\nprint(len(best))\\nset(scores_df.parent.values) - set(best.parent.values)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best = scores_df[scores_df[\"dslf_fa13_X\"] < -1]\n",
    "best = best[best[\"dslf_fa13_Y\"] < -1]\n",
    "best = best[best[\"rmsd_final_X\"] < 0.5]\n",
    "best = best[best[\"rmsd_final_Y\"] < 0.5]\n",
    "\n",
    "idces = list(best.index)\n",
    "\n",
    "print(len(set(best.state.values)))\n",
    "print(len(best))\n",
    "set(scores_df.parent.values) - set(best.parent.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 39;\n",
       "                var nbb_unformatted_code = \"with open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"best_dslf_X.list\\\"), \\\"w\\\") as f:\\n    for i in best.index:\\n        print(best.loc[i][\\\"X_nocys_path\\\"], file=f)\\nwith open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"best_dslf_Y.list\\\"), \\\"w\\\") as f:\\n    for i in best.index:\\n        print(best.loc[i][\\\"Y_nocys_path\\\"], file=f)\";\n",
       "                var nbb_formatted_code = \"with open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"best_dslf_X.list\\\"), \\\"w\\\") as f:\\n    for i in best.index:\\n        print(best.loc[i][\\\"X_nocys_path\\\"], file=f)\\nwith open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"best_dslf_Y.list\\\"), \\\"w\\\") as f:\\n    for i in best.index:\\n        print(best.loc[i][\\\"Y_nocys_path\\\"], file=f)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with open(os.path.join(os.getcwd(), \"06_score_dslf\", \"best_dslf_X.list\"), \"w\") as f:\n",
    "    for i in best.index:\n",
    "        print(best.loc[i][\"X_nocys_path\"], file=f)\n",
    "with open(os.path.join(os.getcwd(), \"06_score_dslf\", \"best_dslf_Y.list\"), \"w\") as f:\n",
    "    for i in best.index:\n",
    "        print(best.loc[i][\"Y_nocys_path\"], file=f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get ok disulfides"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "48\n",
      "1778\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'DHR18_nocys',\n",
       " 'DHR39',\n",
       " 'DHR76',\n",
       " 'DHR79',\n",
       " 'DHR82',\n",
       " 'hDHR14_5CWH',\n",
       " 'hDHR14_5H7C',\n",
       " 'hDHR53_5CWK',\n",
       " 'hDHR64_5CWM',\n",
       " 'hTH_DHR_DN_T6'}"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 41;\n",
       "                var nbb_unformatted_code = \"good = scores_df[scores_df[\\\"dslf_fa13_X\\\"] < -0.9]\\ngood = good[good[\\\"dslf_fa13_Y\\\"] < -0.9]\\ngood = good[good[\\\"rmsd_final_X\\\"] < 0.5]\\ngood = good[good[\\\"rmsd_final_Y\\\"] < 0.5]\\n\\nidces = list(good.index)\\n\\nprint(len(set(good.state.values)))\\nprint(len(good))\\nset(scores_df.parent.values) - set(good.parent.values)\";\n",
       "                var nbb_formatted_code = \"good = scores_df[scores_df[\\\"dslf_fa13_X\\\"] < -0.9]\\ngood = good[good[\\\"dslf_fa13_Y\\\"] < -0.9]\\ngood = good[good[\\\"rmsd_final_X\\\"] < 0.5]\\ngood = good[good[\\\"rmsd_final_Y\\\"] < 0.5]\\n\\nidces = list(good.index)\\n\\nprint(len(set(good.state.values)))\\nprint(len(good))\\nset(scores_df.parent.values) - set(good.parent.values)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "good = scores_df[scores_df[\"dslf_fa13_X\"] < -0.9]\n",
    "good = good[good[\"dslf_fa13_Y\"] < -0.9]\n",
    "good = good[good[\"rmsd_final_X\"] < 0.5]\n",
    "good = good[good[\"rmsd_final_Y\"] < 0.5]\n",
    "\n",
    "idces = list(good.index)\n",
    "\n",
    "print(len(set(good.state.values)))\n",
    "print(len(good))\n",
    "set(scores_df.parent.values) - set(good.parent.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 42;\n",
       "                var nbb_unformatted_code = \"with open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"good_dslf_X.list\\\"), \\\"w\\\") as f:\\n    for i in good.index:\\n        print(good.loc[i][\\\"X_nocys_path\\\"], file=f)\\nwith open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"good_dslf_Y.list\\\"), \\\"w\\\") as f:\\n    for i in good.index:\\n        print(good.loc[i][\\\"Y_nocys_path\\\"], file=f)\";\n",
       "                var nbb_formatted_code = \"with open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"good_dslf_X.list\\\"), \\\"w\\\") as f:\\n    for i in good.index:\\n        print(good.loc[i][\\\"X_nocys_path\\\"], file=f)\\nwith open(os.path.join(os.getcwd(), \\\"06_score_dslf\\\", \\\"good_dslf_Y.list\\\"), \\\"w\\\") as f:\\n    for i in good.index:\\n        print(good.loc[i][\\\"Y_nocys_path\\\"], file=f)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with open(os.path.join(os.getcwd(), \"06_score_dslf\", \"good_dslf_X.list\"), \"w\") as f:\n",
    "    for i in good.index:\n",
    "        print(good.loc[i][\"X_nocys_path\"], file=f)\n",
    "with open(os.path.join(os.getcwd(), \"06_score_dslf\", \"good_dslf_Y.list\"), \"w\") as f:\n",
    "    for i in good.index:\n",
    "        print(good.loc[i][\"Y_nocys_path\"], file=f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Unused blocks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def basename_noext(abspath):\n",
    "    basename = abspath.split(\"/\")[-1]\n",
    "    no_ext = basename.split(\".\")[0]\n",
    "    return no_ext\n",
    "\n",
    "for des in one_trp[one_trp[\"parent\"] == \"DHR82\"].index:\n",
    "    i += 1\n",
    "    print(f\"# {one_trp.loc[des].state}\")\n",
    "    print(f\"load {one_trp.loc[des].X_nocys_path}\")\n",
    "    print(f\"load {one_trp.loc[des].Y_nocys_path}\")\n",
    "    print(\n",
    "        f\"super {basename_noext(one_trp.loc[des].X_nocys_path)} and resi 1-10, {basename_noext(one_trp.loc[des].Y_nocys_path)} and resi 1-10\"\n",
    "    )\n",
    "    print()\n",
    "    \n",
    "print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pyrosetta\n",
    "pyrosetta.distributed.init()\n",
    "import pyrosetta.distributed.io as io\n",
    "from pyrosetta.distributed import cluster\n",
    "\n",
    "final_pdbs = glob(os.path.join(os.getcwd(), \"06_states/*XY.pdb\"))\n",
    "to_inspect = []\n",
    "with open(\"final_to_inspect.list\", \"w+\") as f:\n",
    "    for pdb in final_pdbs:\n",
    "        basename_no_ext = pdb.split(\"_XY\")[0]\n",
    "        decoys = list(no_explode.index)\n",
    "        for decoy in decoys:\n",
    "            decoy_basename = cluster.get_scores_dict(decoy)[\"instance\"][\"tasks\"][\"-s\"].split(\"/\")[-1].replace(\".pdb\", \"\")\n",
    "            if decoy_basename in basename_no_ext:\n",
    "                decoys.remove(decoy)\n",
    "                print(pdb.replace(\"X\", \"\", 1), file=f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "os.makedirs(\"06_to_order\", exist_ok=True)\n",
    "tag = \"GHHHHHHGS\"\n",
    "all_id = {}\n",
    "with open(\"final_to_inspect.list\", \"r\") as f:\n",
    "    with open(\"06_to_order/all.fasta\", \"w+\") as out:\n",
    "        for i, line in enumerate(f):\n",
    "            all_id[f\"pf_{i}\"] = f\"{line.rstrip()}\"\n",
    "            print(f\">pf_{i}\", file=out)\n",
    "            print(tag+io.pose_from_file(line.rstrip()).pose.sequence(), file=out)\n",
    "all_id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "with open(\"06_to_order/all_ids.json\", \"w+\") as f:\n",
    "    print(json.dumps(all_id), file=f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = json.loads(\"06_to_order/all_ids.json\")\n",
    "test\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyrosetta.distributed import cluster\n",
    "with open(\"final_to_inspect.list\", \"r\") as f:\n",
    "    for line in f:\n",
    "        cluster.get_scores_dict(line.rstrip()) #[\"instance\"][\"tasks\"][\"-s\"].split(\"/\")[-1].replace(\".pdb\", \"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "t.scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "%%time\n",
    "import pyrosetta\n",
    "from pyrosetta.distributed import cluster\n",
    "import pyrosetta.distributed.io as io\n",
    "\n",
    "flags = \"\"\"\n",
    "-out:level 300\n",
    "-precompute_ig 1\n",
    "-detect_disulf false\n",
    "-corrections::beta_nov16 true\n",
    "-holes:dalphaball /home/bcov/ppi/tutorial_build/main/source/external/DAlpahBall/DAlphaBall.gcc\n",
    "-indexed_structure_store:fragment_store /net/databases/VALL_clustered/connect_chains/ss_grouped_vall_helix_shortLoop.h5\n",
    "\"\"\"\n",
    "pyrosetta.distributed.init(\" \".join(flags.replace(\"\\n\\t\", \" \").split()))\n",
    "t = almost_linkres(\n",
    "    None,\n",
    "    **{\n",
    "        \"-s\": \"/mnt/home/pleung/projects/bistable_bundle/r4/hinges/05_msd_runs_1/10/0009d2892b7dbae457c1c6f8bc40fe5524115bd1423ebeb4.pdb\",\n",
    "        \"out_path\": \"test\",\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 59 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 97 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 59 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 97 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 59 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 97 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 59 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 97 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 40 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 83 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 40 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 83 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 40 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 83 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 40 BRANCH 1\n",
      "core.pose.util: {0} [ WARNING ] Unable to find atom_tree atom for this Rosetta branch connection angle: residue 83 BRANCH 1\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "core.chemical.AtomICoor: {0} [ WARNING ] IcoorAtomID::atom_id(): Cannot get atom_id for POLYMER_LOWER of residue LEU 1.  Returning BOGUS ID instead.\n",
      "core.conformation.Residue: {0} [ WARNING ] missing an atom: 1  H   that depends on a nonexistent polymer connection!\n",
      "core.conformation.Residue: {0} [ WARNING ]  --> generating it using idealized coordinates.\n",
      "CPU times: user 14min 22s, sys: 991 ms, total: 14min 23s\n",
      "Wall time: 15min 11s\n"
     ]
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "            setTimeout(function() {\n",
       "                var nbb_cell_id = 13;\n",
       "                var nbb_unformatted_code = \"%%time\\nimport pyrosetta\\nfrom pyrosetta.distributed import cluster\\nimport pyrosetta.distributed.io as io\\n\\nflags = \\\"\\\"\\\"\\n-corrections::beta_nov16 true\\n\\\"\\\"\\\"\\npyrosetta.distributed.init(\\\" \\\".join(flags.replace(\\\"\\\\n\\\\t\\\", \\\" \\\").split()))\\nt = score_dslf(\\n    None,\\n    **{\\n        \\\"-s\\\": \\\"/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_1c3348aaff704bdaa73c6d67fa2db82a.pdb.bz2\\\",\\n    }\\n)\";\n",
       "                var nbb_formatted_code = \"%%time\\nimport pyrosetta\\nfrom pyrosetta.distributed import cluster\\nimport pyrosetta.distributed.io as io\\n\\nflags = \\\"\\\"\\\"\\n-corrections::beta_nov16 true\\n\\\"\\\"\\\"\\npyrosetta.distributed.init(\\\" \\\".join(flags.replace(\\\"\\\\n\\\\t\\\", \\\" \\\").split()))\\nt = score_dslf(\\n    None,\\n    **{\\n        \\\"-s\\\": \\\"/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_1c3348aaff704bdaa73c6d67fa2db82a.pdb.bz2\\\",\\n    }\\n)\";\n",
       "                var nbb_cells = Jupyter.notebook.get_cells();\n",
       "                for (var i = 0; i < nbb_cells.length; ++i) {\n",
       "                    if (nbb_cells[i].input_prompt_number == nbb_cell_id) {\n",
       "                        if (nbb_cells[i].get_text() == nbb_unformatted_code) {\n",
       "                             nbb_cells[i].set_text(nbb_formatted_code);\n",
       "                        }\n",
       "                        break;\n",
       "                    }\n",
       "                }\n",
       "            }, 500);\n",
       "            "
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "import pyrosetta\n",
    "from pyrosetta.distributed import cluster\n",
    "import pyrosetta.distributed.io as io\n",
    "\n",
    "flags = \"\"\"\n",
    "-corrections::beta_nov16 true\n",
    "\"\"\"\n",
    "pyrosetta.distributed.init(\" \".join(flags.replace(\"\\n\\t\", \" \").split()))\n",
    "t = score_dslf(\n",
    "    None,\n",
    "    **{\n",
    "        \"-s\": \"/mnt/home/pleung/projects/bistable_bundle/r4/hinges/06_cleanup/decoys/0000/2021.06.26.14.24.46.542335_1c3348aaff704bdaa73c6d67fa2db82a.pdb.bz2\",\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "phil (3.8.2)",
   "language": "python",
   "name": "phil"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}