Source code for mbuild.compound

__all__ = ['load', 'clone', 'Compound', 'Particle']

from collections import OrderedDict, defaultdict, Iterable
from copy import deepcopy
import itertools
import os
import sys
import tempfile
from warnings import warn

import mdtraj as md
from mdtraj.core.element import get_by_symbol
import numpy as np
from oset import oset as OrderedSet
import parmed as pmd
from parmed.periodic_table import AtomicNum, element_by_name, Mass, Element

from mbuild.bond_graph import BondGraph
from mbuild.box import Box
from mbuild.exceptions import MBuildError
from mbuild.utils.decorators import deprecated
from mbuild.formats.xyz import read_xyz, write_xyz
from mbuild.formats.json_formats import compound_to_json, compound_from_json
from mbuild.formats.hoomdxml import write_hoomdxml
from mbuild.formats.lammpsdata import write_lammpsdata
from mbuild.formats.gsdwriter import write_gsd
from mbuild.formats.par_writer import write_par
from mbuild.periodic_kdtree import PeriodicCKDTree
from mbuild.utils.io import run_from_ipython, import_, has_networkx
from mbuild.utils.jsutils import overwrite_nglview_default
from mbuild.coordinate_transform import _translate, _rotate


def load(filename_or_object, relative_to_module=None, compound=None, coords_only=False,
         rigid=False, use_parmed=False, smiles=False, 
         infer_hierarchy=True, **kwargs):
    """Load a file or an existing topology into an mbuild compound.

    Files are read using the MDTraj package unless the `use_parmed` argument is
    specified as True. Please refer to http://mdtraj.org/1.8.0/load_functions.html
    for formats supported by MDTraj and https://parmed.github.io/ParmEd/html/
    readwrite.html for formats supported by ParmEd.

    Parameters
    ----------
    filename_or_object : str, mdtraj.Trajectory, parmed.Structure, mbuild.Compound,
            pybel.Molecule
        Name of the file or topology from which to load atom and bond information.
    relative_to_module : str, optional, default=None
        Instead of looking in the current working directory, look for the file
        where this module is defined. This is typically used in Compound
        classes that will be instantiated from a different directory
        (such as the Compounds located in mbuild.lib).
    compound : mb.Compound, optional, default=None
        Existing compound to load atom and bond information into.
    coords_only : bool, optional, default=False
        Only load the coordinates into an existing compound.
    rigid : bool, optional, default=False
        Treat the compound as a rigid body
    use_parmed : bool, optional, default=False
        Use readers from ParmEd instead of MDTraj.
    smiles: bool, optional, default=False
        Use Open Babel to parse filename as a SMILES string
        or file containing a SMILES string.
    infer_hierarchy : bool, optional, default=True
        If True, infer hierarchy from chains and residues
    **kwargs : keyword arguments
        Key word arguments passed to mdTraj for loading.

    Returns
    -------
    compound : mb.Compound

    """
    # If compound doesn't exist, we will initialize one
    if compound is None:
        compound = Compound()

    # First check if we are loading from an existing parmed or trajectory structure
    type_dict = {
        pmd.Structure:compound.from_parmed,
        md.Trajectory:compound.from_trajectory,
    }
    try:
        pybel = import_('pybel')
        type_dict.update({pybel.Molecule:compound.from_pybel})
    except ImportError:
        pass

    if isinstance(filename_or_object, Compound):
        return filename_or_object
    for type in type_dict:
        if isinstance(filename_or_object, type):
            type_dict[type](filename_or_object,coords_only=coords_only, 
                    infer_hierarchy=infer_hierarchy, **kwargs)
            return compound
    if not isinstance(filename_or_object, str):
        raise ValueError('Input not supported.')

    # Handle mbuild *.py files containing a class that wraps a structure file
    # in its own folder. E.g., you build a system from ~/foo.py and it imports
    # from ~/bar/baz.py where baz.py loads ~/bar/baz.pdb.
    if relative_to_module:
        script_path = os.path.realpath(
            sys.modules[relative_to_module].__file__)
        file_dir = os.path.dirname(script_path)
        filename_or_object = os.path.join(file_dir, filename_or_object)

    # Handle the case of a xyz and json file, which must use an internal reader
    extension = os.path.splitext(filename_or_object)[-1]
    if extension == '.json':
        compound = compound_from_json(filename_or_object)
        return compound

    if extension == '.xyz' and not 'top' in kwargs:
        if coords_only:
            tmp = read_xyz(filename_or_object)
            if tmp.n_particles != compound.n_particles:
                raise ValueError('Number of atoms in {filename_or_object} does not match'
                                 ' {compound}'.format(**locals()))
            ref_and_compound = zip(tmp._particles(include_ports=False),
                                   compound.particles(include_ports=False))
            for ref_particle, particle in ref_and_compound:
                particle.pos = ref_particle.pos
        else:
            compound = read_xyz(filename_or_object, compound=compound)
        return compound

    if extension == '.sdf':
        pybel = import_('pybel')
        pybel_mol = pybel.readfile('sdf', filename_or_object)
        # pybel returns a generator, so we grab the first molecule of a list of len 1
        # Raise ValueError user if there are more molecules
        pybel_mol = [i for i in pybel_mol]
        if len(pybel_mol) == 1:
            compound.from_pybel(pybel_mol[0])
        else:
            compound.from_pybel(pybel_mol[0])
            raise ValueError("More than one pybel molecule in file, more than one pybel "
                 "molecule is not supported, using {}".format(filename_or_object))
        return compound

    if use_parmed:
        warn(
            "use_parmed set to True.  Bonds may be inferred from inter-particle "
            "distances and standard residue templates!")
        structure = pmd.load_file(filename_or_object, structure=True, **kwargs)
        compound.from_parmed(structure, coords_only=coords_only, 
                infer_hierarchy=infer_hierarchy)

    elif smiles:
        pybel = import_('pybel')
        # First we try treating filename_or_object as a SMILES string
        try:
            mymol = pybel.readstring("smi", filename_or_object)
        # Now we treat it as a filename
        except(OSError, IOError):
            # For now, we only support reading in a single smiles molecule,
            # but pybel returns a generator, so we get the first molecule
            # and warn the user if there is more

            mymol_generator = pybel.readfile("smi", filename_or_object)
            mymol_list = list(mymol_generator)
            if len(mymol_list) == 1:
                mymol = mymol_list[0]
            else:
                mymol = mymol_list[0]
                warn("More than one SMILES string in file, more than one SMILES "
                     "string is not supported, using {}".format(mymol.write("smi")))

        # We create a temporary directory and mol2 file that will created from the smiles string
        # A ParmEd structure and subsequenc mBuild compound will be created from this mol2 file
        tmp_dir = tempfile.mkdtemp()
        temp_file = os.path.join(tmp_dir, 'smiles_to_mol2_intermediate.mol2')
        mymol.make3D()
        compound = Compound()
        compound.from_pybel(mymol, infer_hierarchy=infer_hierarchy, ignore_box_warn=True)

    else:
        traj = md.load(filename_or_object, **kwargs)
        compound.from_trajectory(traj, frame=-1, coords_only=coords_only,
                infer_hierarchy=infer_hierarchy)

    if rigid:
        compound.label_rigid_bodies()
    return compound


def clone(existing_compound, clone_of=None, root_container=None):
    """A faster alternative to deepcopying.

    Does not resolve circular dependencies. This should be safe provided
    you never try to add the top of a Compound hierarchy to a
    sub-Compound.

    Parameters
    ----------
    existing_compound : mb.Compound
        Existing Compound that will be copied

    Other Parameters
    ----------------
    clone_of : dict, optional
    root_container : mb.Compound, optional

    """
    if clone_of is None:
        clone_of = dict()

    newone = existing_compound._clone(clone_of=clone_of,
                                      root_container=root_container)
    existing_compound._clone_bonds(clone_of=clone_of)
    return newone


[docs]class Compound(object): """A building block in the mBuild hierarchy. Compound is the superclass of all composite building blocks in the mBuild hierarchy. That is, all composite building blocks must inherit from compound, either directly or indirectly. The design of Compound follows the Composite design pattern (Gamma, Erich; Richard Helm; Ralph Johnson; John M. Vlissides (1995). Design Patterns: Elements of Reusable Object-Oriented Software. Addison-Wesley. p. 395. ISBN 0-201-63361-2.), with Compound being the composite, and Particle playing the role of the primitive (leaf) part, where Particle is in fact simply an alias to the Compound class. Compound maintains a list of children (other Compounds contained within), and provides a means to tag the children with labels, so that the compounds can be easily looked up later. Labels may also point to objects outside the Compound's containment hierarchy. Compound has built-in support for copying and deepcopying Compound hierarchies, enumerating particles or bonds in the hierarchy, proximity based searches, visualization, I/O operations, and a number of other convenience methods. Parameters ---------- subcompounds : mb.Compound or list of mb.Compound, optional, default=None One or more compounds to be added to self. name : str, optional, default=self.__class__.__name__ The type of Compound. pos : np.ndarray, shape=(3,), dtype=float, optional, default=[0, 0, 0] The position of the Compound in Cartestian space charge : float, optional, default=0.0 Currently not used. Likely removed in next release. periodicity : np.ndarray, shape=(3,), dtype=float, optional, default=[0, 0, 0] The periodic lengths of the Compound in the x, y and z directions. Defaults to zeros which is treated as non-periodic. port_particle : bool, optional, default=False Whether or not this Compound is part of a Port box : mb.Box, optional The simulation box containing the compound. Also accounts for the periodicity. Defaults to None which is treated as non-periodic. Attributes ---------- bond_graph : mb.BondGraph Graph-like object that stores bond information for this Compound children : OrderedSet Contains all children (other Compounds). labels : OrderedDict Labels to Compound/Atom mappings. These do not necessarily need not be in self.children. parent : mb.Compound The parent Compound that contains this part. Can be None if this compound is the root of the containment hierarchy. referrers : set Other compounds that reference this part with labels. rigid_id : int, default=None The ID of the rigid body that this Compound belongs to. Only Particles (the bottom of the containment hierarchy) can have integer values for `rigid_id`. Compounds containing rigid particles will always have `rigid_id == None`. See also `contains_rigid`. boundingbox : mb.Box The bounds (xmin, xmax, ymin, ymax, zmin, zmax) of particles in Compound center contains_rigid max_rigid_id n_particles n_bonds root xyz xyz_with_ports """ def __init__(self, subcompounds=None, name=None, pos=None, charge=0.0, periodicity=None, box=None, port_particle=False): super(Compound, self).__init__() if name: if not isinstance(name, str): raise ValueError( 'Compound.name should be a string. You passed ' '{}'.format(name)) self.name = name else: self.name = self.__class__.__name__ # A periodicity of zero in any direction is treated as non-periodic. if periodicity is None: self._periodicity = np.array([0.0, 0.0, 0.0]) else: self._periodicity = np.asarray(periodicity) if pos is not None: self._pos = np.asarray(pos, dtype=float) else: self._pos = np.zeros(3) self.parent = None self.children = OrderedSet() self.labels = OrderedDict() self.referrers = set() self.bond_graph = None self.port_particle = port_particle self._rigid_id = None self._contains_rigid = False self._check_if_contains_rigid_bodies = False self.box = box # self.add() must be called after labels and children are initialized. if subcompounds: if charge: raise MBuildError( 'Cannot set the charge of a Compound containing ' 'subcompounds.') self.add(subcompounds) self._charge = 0.0 else: self._charge = charge
[docs] def particles(self, include_ports=False): """Return all Particles of the Compound. Parameters ---------- include_ports : bool, optional, default=False Include port particles Yields ------- mb.Compound The next Particle in the Compound """ if not self.children: yield self else: for particle in self._particles(include_ports): yield particle
def _particles(self, include_ports=False): """Return all Particles of the Compound. """ for child in self.successors(): if not child.children: if include_ports or not child.port_particle: yield child
[docs] def successors(self): """Yield Compounds below self in the hierarchy. Yields ------- mb.Compound The next Particle below self in the hierarchy """ if not self.children: return for part in self.children: # Parts local to the current Compound. yield part # Parts further down the hierarchy. for subpart in part.successors(): yield subpart
@property def n_particles(self): """Return the number of Particles in the Compound. Returns ------- int The number of Particles in the Compound """ if not self.children: return 1 else: return self._n_particles(include_ports=False) def _n_particles(self, include_ports=False): """Return the number of Particles in the Compound. """ return sum(1 for _ in self._particles(include_ports)) def _contains_only_ports(self): for part in self.children: if not part.port_particle: return False return True
[docs] def ancestors(self): """Generate all ancestors of the Compound recursively. Yields ------ mb.Compound The next Compound above self in the hierarchy """ if self.parent is not None: yield self.parent for ancestor in self.parent.ancestors(): yield ancestor
@property def root(self): """The Compound at the top of self's hierarchy. Returns ------- mb.Compound The Compound at the top of self's hierarchy """ parent = None for parent in self.ancestors(): pass if parent is None: return self return parent
[docs] def particles_by_name(self, name): """Return all Particles of the Compound with a specific name Parameters ---------- name : str Only particles with this name are returned Yields ------ mb.Compound The next Particle in the Compound with the user-specified name """ for particle in self.particles(): if particle.name == name: yield particle
@property def charge(self): return sum([particle._charge for particle in self.particles()]) @charge.setter def charge(self, value): if self._contains_only_ports(): self._charge = value else: raise AttributeError( "charge is immutable for Compounds that are " "not at the bottom of the containment hierarchy.") @property def rigid_id(self): return self._rigid_id @rigid_id.setter def rigid_id(self, value): if self._contains_only_ports(): self._rigid_id = value for ancestor in self.ancestors(): ancestor._check_if_contains_rigid_bodies = True else: raise AttributeError( "rigid_id is immutable for Compounds that are " "not at the bottom of the containment hierarchy.") @property def contains_rigid(self): """Returns True if the Compound contains rigid bodies If the Compound contains any particle with a rigid_id != None then contains_rigid will return True. If the Compound has no children (i.e. the Compound resides at the bottom of the containment hierarchy) then contains_rigid will return False. Returns ------- bool True if the Compound contains any particle with a rigid_id != None Notes ----- The private variable '_check_if_contains_rigid_bodies' is used to help cache the status of 'contains_rigid'. If '_check_if_contains_rigid_bodies' is False, then the rigid body containment of the Compound has not changed, and the particle tree is not traversed, boosting performance. """ if self._check_if_contains_rigid_bodies: self._check_if_contains_rigid_bodies = False if any(particle.rigid_id is not None for particle in self._particles()): self._contains_rigid = True else: self._contains_rigid = False return self._contains_rigid @property def max_rigid_id(self): """Returns the maximum rigid body ID contained in the Compound. This is usually used by compound.root to determine the maximum rigid_id in the containment hierarchy. Returns ------- int or None The maximum rigid body ID contained in the Compound. If no rigid body IDs are found, None is returned """ try: return max([particle.rigid_id for particle in self.particles() if particle.rigid_id is not None]) except ValueError: return
[docs] def rigid_particles(self, rigid_id=None): """Generate all particles in rigid bodies. If a rigid_id is specified, then this function will only yield particles with a matching rigid_id. Parameters ---------- rigid_id : int, optional Include only particles with this rigid body ID Yields ------ mb.Compound The next particle with a rigid_id that is not None, or the next particle with a matching rigid_id if specified """ for particle in self.particles(): if rigid_id is not None: if particle.rigid_id == rigid_id: yield particle else: if particle.rigid_id is not None: yield particle
[docs] def label_rigid_bodies(self, discrete_bodies=None, rigid_particles=None): """Designate which Compounds should be treated as rigid bodies If no arguments are provided, this function will treat the compound as a single rigid body by providing all particles in `self` with the same rigid_id. If `discrete_bodies` is not None, each instance of a Compound with a name found in `discrete_bodies` will be treated as a unique rigid body. If `rigid_particles` is not None, only Particles (Compounds at the bottom of the containment hierarchy) matching this name will be considered part of the rigid body. Parameters ---------- discrete_bodies : str or list of str, optional, default=None Name(s) of Compound instances to be treated as unique rigid bodies. Compound instances matching this (these) name(s) will be provided with unique rigid_ids rigid_particles : str or list of str, optional, default=None Name(s) of Compound instances at the bottom of the containment hierarchy (Particles) to be included in rigid bodies. Only Particles matching this (these) name(s) will have their rigid_ids altered to match the rigid body number. Examples -------- Creating a rigid benzene >>> import mbuild as mb >>> from mbuild.utils.io import get_fn >>> benzene = mb.load(get_fn('benzene.mol2')) >>> benzene.label_rigid_bodies() Creating a semi-rigid benzene, where only the carbons are treated as a rigid body >>> import mbuild as mb >>> from mbuild.utils.io import get_fn >>> benzene = mb.load(get_fn('benzene.mol2')) >>> benzene.label_rigid_bodies(rigid_particles='C') Create a box of rigid benzenes, where each benzene has a unique rigid body ID. >>> import mbuild as mb >>> from mbuild.utils.io import get_fn >>> benzene = mb.load(get_fn('benzene.mol2')) >>> benzene.name = 'Benzene' >>> filled = mb.fill_box(benzene, ... n_compounds=10, ... box=[0, 0, 0, 4, 4, 4]) >>> filled.label_rigid_bodies(distinct_bodies='Benzene') Create a box of semi-rigid benzenes, where each benzene has a unique rigid body ID and only the carbon portion is treated as rigid. >>> import mbuild as mb >>> from mbuild.utils.io import get_fn >>> benzene = mb.load(get_fn('benzene.mol2')) >>> benzene.name = 'Benzene' >>> filled = mb.fill_box(benzene, ... n_compounds=10, ... box=[0, 0, 0, 4, 4, 4]) >>> filled.label_rigid_bodies(distinct_bodies='Benzene', ... rigid_particles='C') """ if discrete_bodies is not None: if isinstance(discrete_bodies, str): discrete_bodies = [discrete_bodies] if rigid_particles is not None: if isinstance(rigid_particles, str): rigid_particles = [rigid_particles] if self.root.max_rigid_id is not None: rigid_id = self.root.max_rigid_id + 1 warn("{} rigid bodies already exist. Incrementing 'rigid_id'" "starting from {}.".format(rigid_id, rigid_id)) else: rigid_id = 0 for successor in self.successors(): if discrete_bodies and successor.name not in discrete_bodies: continue for particle in successor.particles(): if rigid_particles and particle.name not in rigid_particles: continue particle.rigid_id = rigid_id if discrete_bodies: rigid_id += 1
[docs] def unlabel_rigid_bodies(self): """Remove all rigid body labels from the Compound """ self._check_if_contains_rigid_bodies = True for child in self.children: child._check_if_contains_rigid_bodies = True for particle in self.particles(): particle.rigid_id = None
def _increment_rigid_ids(self, increment): """Increment the rigid_id of all rigid Particles in a Compound Adds `increment` to the rigid_id of all Particles in `self` that already have an integer rigid_id. """ for particle in self.particles(): if particle.rigid_id is not None: particle.rigid_id += increment def _reorder_rigid_ids(self): """Reorder rigid body IDs ensuring consecutiveness. Primarily used internally to ensure consecutive rigid_ids following removal of a Compound. """ max_rigid = self.max_rigid_id unique_rigid_ids = sorted( set([p.rigid_id for p in self.rigid_particles()])) n_unique_rigid = len(unique_rigid_ids) if max_rigid and n_unique_rigid != max_rigid + 1: missing_rigid_id = ( unique_rigid_ids[-1] * (unique_rigid_ids[-1] + 1)) / 2 - sum(unique_rigid_ids) for successor in self.successors(): if successor.rigid_id is not None: if successor.rigid_id > missing_rigid_id: successor.rigid_id -= 1 if self.rigid_id: if self.rigid_id > missing_rigid_id: self.rigid_id -= 1
[docs] def add(self, new_child, label=None, containment=True, replace=False, inherit_periodicity=True, inherit_box=False, reset_rigid_ids=True): """Add a part to the Compound. Note: This does not necessarily add the part to self.children but may instead be used to add a reference to the part to self.labels. See 'containment' argument. Parameters ---------- new_child : mb.Compound or list-like of mb.Compound The object(s) to be added to this Compound. label : str, optional A descriptive string for the part. containment : bool, optional, default=True Add the part to self.children. replace : bool, optional, default=True Replace the label if it already exists. inherit_periodicity : bool, optional, default=True Replace the periodicity of self with the periodicity of the Compound being added inherit_box: bool, optional, default=False Replace the box of self with the box of the Compound being added reset_rigid_ids : bool, optional, default=True If the Compound to be added contains rigid bodies, reset the rigid_ids such that values remain distinct from rigid_ids already present in `self`. Can be set to False if attempting to add Compounds to an existing rigid body. """ # Support batch add via lists, tuples and sets. if (isinstance(new_child, Iterable) and not isinstance(new_child, str)): for child in new_child: self.add(child, reset_rigid_ids=reset_rigid_ids) return if not isinstance(new_child, Compound): raise ValueError('Only objects that inherit from mbuild.Compound ' 'can be added to Compounds. You tried to add ' '"{}".'.format(new_child)) if new_child.contains_rigid or new_child.rigid_id is not None: if self.contains_rigid and reset_rigid_ids: new_child._increment_rigid_ids(increment=self.max_rigid_id + 1) self._check_if_contains_rigid_bodies = True if self.rigid_id is not None: self.rigid_id = None # Create children and labels on the first add operation if self.children is None: self.children = OrderedSet() if self.labels is None: self.labels = OrderedDict() if containment: if new_child.parent is not None: raise MBuildError('Part {} already has a parent: {}'.format( new_child, new_child.parent)) self.children.add(new_child) new_child.parent = self if new_child.bond_graph is not None: if self.root.bond_graph is None: self.root.bond_graph = new_child.bond_graph else: self.root.bond_graph.compose(new_child.bond_graph) new_child.bond_graph = None # Add new_part to labels. Does not currently support batch add. if label is None: label = '{0}[$]'.format(new_child.__class__.__name__) if label.endswith('[$]'): label = label[:-3] if label not in self.labels: self.labels[label] = [] label_pattern = label + '[{}]' count = len(self.labels[label]) self.labels[label].append(new_child) label = label_pattern.format(count) if not replace and label in self.labels: raise MBuildError('Label "{0}" already exists in {1}.'.format( label, self)) else: self.labels[label] = new_child new_child.referrers.add(self) if (inherit_periodicity and isinstance(new_child, Compound) and new_child.periodicity.any()): self.periodicity = new_child.periodicity # If parent has no box --> inherit child box # If parent has box --> keep unless inherit_box == True # If inherit_box == True, parent box != None, child_box == None, # keep parent box anyway and warn # If inherit_box == False, child.box != None, warn if self.box is None: if new_child.box is not None: self.box = new_child.box else: if inherit_box: if new_child.box is None: warn( "The Compound you are adding has no box but " "inherit_box=True. The box of the original " "Compound will remain unchanged." ) else: self.box = new_child.box else: if new_child.box is not None: warn( "The Compound you are adding has a box. " "The box of the parent compound will be used. Use " "inherit_box = True if you wish to replace the parent " "compound box with that of Compound being added." ) # Check that bounding box is within box after adding compound if self.box: if (self.box.lengths < self.boundingbox.lengths).any(): warn( "After adding new Compound, Compound.box.lengths < " "Compound.boundingbox.lengths. There may be particles " "outside of the defined simulation box" )
[docs] def remove(self, objs_to_remove): """ Cleanly remove children from the Compound. Parameters ---------- objs_to_remove : mb.Compound or list of mb.Compound The Compound(s) to be removed from self """ # Preprocessing and validating input type from mbuild.port import Port if not hasattr(objs_to_remove, '__iter__'): objs_to_remove = [objs_to_remove] objs_to_remove = set(objs_to_remove) # If nothing is to be remove, do nothing if len(objs_to_remove) == 0: return # Remove Port objects separately ports_removed = set() for obj in objs_to_remove: if isinstance(obj, Port): ports_removed.add(obj) self._remove(obj) obj.parent.children.remove(obj) self._remove_references(obj) objs_to_remove = objs_to_remove - ports_removed # Get particles to remove particles_to_remove = set([particle for obj in objs_to_remove for particle in obj.particles()]) # Recursively get container compounds to remove to_remove = list() def _check_if_empty(child): if child in to_remove: return if set(child.particles()).issubset(particles_to_remove): if child.parent: to_remove.append(child) _check_if_empty(child.parent) else: warn("This will remove all particles in " "compound {}".format(self)) return for particle in particles_to_remove: _check_if_empty(particle) # Fix rigid_ids and remove obj from bondgraph for removed_part in to_remove: self._remove(removed_part) # Remove references to object for removed_part in to_remove: if removed_part.parent is not None: removed_part.parent.children.remove(removed_part) self._remove_references(removed_part) # Remove ghost ports all_ports_list = list(self.all_ports()) for port in all_ports_list: if port.anchor not in [i for i in self.particles()]: port.parent.children.remove(port) # Check and reorder rigid id for _ in particles_to_remove: if self.contains_rigid: self.root._reorder_rigid_ids()
def _remove(self, removed_part): """Worker for remove(). Fixes rigid IDs and removes bonds""" if removed_part.rigid_id is not None: for ancestor in removed_part.ancestors(): ancestor._check_if_contains_rigid_bodies = True if self.root.bond_graph and self.root.bond_graph.has_node( removed_part): for neighbor in self.root.bond_graph.neighbors( removed_part): self.root.remove_bond((removed_part, neighbor)) self.root.bond_graph.remove_node(removed_part) def _remove_references(self, removed_part): """Remove labels pointing to this part and vice versa. """ removed_part.parent = None # Remove labels in the hierarchy pointing to this part. referrers_to_remove = set() for referrer in removed_part.referrers: if removed_part not in referrer.ancestors(): for label, referred_part in list(referrer.labels.items()): if referred_part is removed_part: del referrer.labels[label] referrers_to_remove.add(referrer) removed_part.referrers -= referrers_to_remove # Remove labels in this part pointing into the hierarchy. labels_to_delete = [] if isinstance(removed_part, Compound): for label, part in list(removed_part.labels.items()): if not isinstance(part, Compound): for p in part: self._remove_references(p) elif removed_part not in part.ancestors(): try: part.referrers.discard(removed_part) except KeyError: pass else: labels_to_delete.append(label) for label in labels_to_delete: removed_part.labels.pop(label, None)
[docs] def referenced_ports(self): """Return all Ports referenced by this Compound. Returns ------- list of mb.Compound A list of all ports referenced by the Compound """ from mbuild.port import Port return [port for port in self.labels.values() if isinstance(port, Port)]
[docs] def all_ports(self): """Return all Ports referenced by this Compound and its successors Returns ------- list of mb.Compound A list of all Ports referenced by this Compound and its successors """ from mbuild.port import Port return [successor for successor in self.successors() if isinstance(successor, Port)]
[docs] def available_ports(self): """Return all unoccupied Ports referenced by this Compound. Returns ------- list of mb.Compound A list of all unoccupied ports referenced by the Compound """ from mbuild.port import Port return [port for port in self.labels.values() if isinstance(port, Port) and not port.used]
[docs] def bonds(self): """Return all bonds in the Compound and sub-Compounds. Yields ------- tuple of mb.Compound The next bond in the Compound See Also -------- bond_graph.edges_iter : Iterates over all edges in a BondGraph """ if self.root.bond_graph: if self.root == self: return self.root.bond_graph.edges_iter() else: return self.root.bond_graph.subgraph( self.particles()).edges_iter() else: return iter(())
@property def n_bonds(self): """Return the number of bonds in the Compound. Returns ------- int The number of bonds in the Compound """ return sum(1 for _ in self.bonds())
[docs] def add_bond(self, particle_pair): """Add a bond between two Particles. Parameters ---------- particle_pair : indexable object, length=2, dtype=mb.Compound The pair of Particles to add a bond between """ if self.root.bond_graph is None: self.root.bond_graph = BondGraph() self.root.bond_graph.add_edge(particle_pair[0], particle_pair[1])
[docs] def generate_bonds(self, name_a, name_b, dmin, dmax): """Add Bonds between all pairs of types a/b within [dmin, dmax]. Parameters ---------- name_a : str The name of one of the Particles to be in each bond name_b : str The name of the other Particle to be in each bond dmin : float The minimum distance between Particles for considering a bond dmax : float The maximum distance between Particles for considering a bond """ particle_kdtree = PeriodicCKDTree( data=self.xyz, bounds=self.periodicity) particle_array = np.array(list(self.particles())) added_bonds = list() for p1 in self.particles_by_name(name_a): nearest = self.particles_in_range(p1, dmax, max_particles=20, particle_kdtree=particle_kdtree, particle_array=particle_array) for p2 in nearest: if p2 == p1: continue bond_tuple = (p1, p2) if id(p1) < id(p2) else (p2, p1) if bond_tuple in added_bonds: continue min_dist = self.min_periodic_distance(p2.pos, p1.pos) if (p2.name == name_b) and (dmin <= min_dist <= dmax): self.add_bond((p1, p2)) added_bonds.append(bond_tuple)
[docs] def remove_bond(self, particle_pair): """Deletes a bond between a pair of Particles Parameters ---------- particle_pair : indexable object, length=2, dtype=mb.Compound The pair of Particles to remove the bond between """ from mbuild.port import Port if self.root.bond_graph is None or not self.root.bond_graph.has_edge( *particle_pair): warn("Bond between {} and {} doesn't exist!".format(*particle_pair)) return self.root.bond_graph.remove_edge(*particle_pair) bond_vector = particle_pair[0].pos - particle_pair[1].pos if np.allclose(bond_vector, np.zeros(3)): warn("Particles {} and {} overlap! Ports will not be added." "".format(*particle_pair)) return distance = np.linalg.norm(bond_vector) particle_pair[0].parent.add(Port(anchor=particle_pair[0], orientation=-bond_vector, separation=distance / 2), 'port[$]') particle_pair[1].parent.add(Port(anchor=particle_pair[1], orientation=bond_vector, separation=distance / 2), 'port[$]')
@property def pos(self): if not self.children: return self._pos else: return self.center @pos.setter def pos(self, value): if not self.children: self._pos = value else: raise MBuildError('Cannot set position on a Compound that has' ' children.') @property def periodicity(self): return self._periodicity @periodicity.setter def periodicity(self, periods): self._periodicity = np.array(periods) @property def box(self): return self._box @box.setter def box(self, box): if box is not None and type(box) != Box: raise TypeError("box must be specified as an mbuild.Box") if self.port_particle and box is not None: raise ValueError("Ports cannot have a box") # TODO: Fix this for non-orthogonal boxes # Make sure the box is bigger than the bounding box if box is not None: if (box.lengths < self.boundingbox.lengths).any(): warn( "Compound.box.lengths < Compound.boundingbox.lengths. " "There may be particles outside of the defined " "simulation box." ) self._box = box @property def xyz(self): """Return all particle coordinates in this compound. Returns ------- pos : np.ndarray, shape=(n, 3), dtype=float Array with the positions of all particles. """ if not self.children: pos = np.expand_dims(self._pos, axis=0) else: arr = np.fromiter(itertools.chain.from_iterable( particle.pos for particle in self.particles()), dtype=float) pos = arr.reshape((-1, 3)) return pos @property def xyz_with_ports(self): """Return all particle coordinates in this compound including ports. Returns ------- pos : np.ndarray, shape=(n, 3), dtype=float Array with the positions of all particles and ports. """ if not self.children: pos = self._pos else: arr = np.fromiter( itertools.chain.from_iterable( particle.pos for particle in self.particles( include_ports=True)), dtype=float) pos = arr.reshape((-1, 3)) return pos @xyz.setter def xyz(self, arrnx3): """Set the positions of the particles in the Compound, excluding the Ports. This function does not set the position of the ports. Parameters ---------- arrnx3 : np.ndarray, shape=(n,3), dtype=float The new particle positions """ if not self.children: if not arrnx3.shape[0] == 1: raise ValueError( 'Trying to set position of {} with more than one' 'coordinate: {}'.format( self, arrnx3)) self.pos = np.squeeze(arrnx3) else: for atom, coords in zip( self._particles( include_ports=False), arrnx3): atom.pos = coords @xyz_with_ports.setter def xyz_with_ports(self, arrnx3): """Set the positions of the particles in the Compound, including the Ports. Parameters ---------- arrnx3 : np.ndarray, shape=(n,3), dtype=float The new particle positions """ if not self.children: if not arrnx3.shape[0] == 1: raise ValueError( 'Trying to set position of {} with more than one' 'coordinate: {}'.format( self, arrnx3)) self.pos = np.squeeze(arrnx3) else: for atom, coords in zip( self._particles( include_ports=True), arrnx3): atom.pos = coords @property def center(self): """The cartesian center of the Compound based on its Particles. Returns ------- np.ndarray, shape=(3,), dtype=float The cartesian center of the Compound based on its Particles """ if np.all(np.isfinite(self.xyz)): return np.mean(self.xyz, axis=0) @property def boundingbox(self): """Compute the bounding box of the compound. Returns ------- mb.Box The bounding box for this Compound """ xyz = self.xyz return Box(mins=xyz.min(axis=0), maxs=xyz.max(axis=0))
[docs] def min_periodic_distance(self, xyz0, xyz1): """Vectorized distance calculation considering minimum image. Parameters ---------- xyz0 : np.ndarray, shape=(3,), dtype=float Coordinates of first point xyz1 : np.ndarray, shape=(3,), dtype=float Coordinates of second point Returns ------- float Vectorized distance between the two points following minimum image convention """ d = np.abs(xyz0 - xyz1) d = np.where(d > 0.5 * self.periodicity, self.periodicity - d, d) return np.sqrt((d ** 2).sum(axis=-1))
[docs] def particles_in_range( self, compound, dmax, max_particles=20, particle_kdtree=None, particle_array=None): """Find particles within a specified range of another particle. Parameters ---------- compound : mb.Compound Reference particle to find other particles in range of dmax : float Maximum distance from 'compound' to look for Particles max_particles : int, optional, default=20 Maximum number of Particles to return particle_kdtree : mb.PeriodicCKDTree, optional KD-tree for looking up nearest neighbors. If not provided, a KD- tree will be generated from all Particles in self particle_array : np.ndarray, shape=(n,), dtype=mb.Compound, optional Array of possible particles to consider for return. If not provided, this defaults to all Particles in self Returns ------- np.ndarray, shape=(n,), dtype=mb.Compound Particles in range of compound according to user-defined limits See Also -------- periodic_kdtree.PerioidicCKDTree : mBuild implementation of kd-trees scipy.spatial.ckdtree : Further details on kd-trees """ if particle_kdtree is None: particle_kdtree = PeriodicCKDTree( data=self.xyz, bounds=self.periodicity) _, idxs = particle_kdtree.query( compound.pos, k=max_particles, distance_upper_bound=dmax) idxs = idxs[idxs != self.n_particles] if particle_array is None: particle_array = np.array(list(self.particles())) return particle_array[idxs]
[docs] def visualize(self, show_ports=False, backend='py3dmol', color_scheme={}): # pragma: no cover """Visualize the Compound using py3dmol (default) or nglview. Allows for visualization of a Compound within a Jupyter Notebook. Parameters ---------- show_ports : bool, optional, default=False Visualize Ports in addition to Particles backend : str, optional, default='py3dmol' Specify the backend package to visualize compounds Currently supported: py3dmol, nglview color_scheme : dict, optional Specify coloring for non-elemental particles keys are strings of the particle names values are strings of the colors i.e. {'_CGBEAD': 'blue'} """ viz_pkg = {'nglview': self._visualize_nglview, 'py3dmol': self._visualize_py3dmol} if run_from_ipython(): if backend.lower() in viz_pkg: return viz_pkg[backend.lower()](show_ports=show_ports, color_scheme=color_scheme) else: raise RuntimeError("Unsupported visualization " + "backend ({}). ".format(backend) + "Currently supported backends include nglview and py3dmol") else: raise RuntimeError('Visualization is only supported in Jupyter ' 'Notebooks.')
def _visualize_py3dmol(self, show_ports=False, color_scheme={}): """Visualize the Compound using py3Dmol. Allows for visualization of a Compound within a Jupyter Notebook. Parameters ---------- show_ports : bool, optional, default=False Visualize Ports in addition to Particles color_scheme : dict, optional Specify coloring for non-elemental particles keys are strings of the particle names values are strings of the colors i.e. {'_CGBEAD': 'blue'} Returns ------ view : py3Dmol.view """ py3Dmol = import_('py3Dmol') cloned = clone(self) modified_color_scheme = {} for name, color in color_scheme.items(): # Py3dmol does some element string conversions, # first character is as-is, rest of the characters are lowercase new_name = name[0] + name[1:].lower() modified_color_scheme[new_name] = color modified_color_scheme[name] = color for particle in cloned.particles(): if not particle.name: particle.name = 'UNK' tmp_dir = tempfile.mkdtemp() cloned.save(os.path.join(tmp_dir, 'tmp.mol2'), show_ports=show_ports, overwrite=True) view = py3Dmol.view() with open(os.path.join(tmp_dir, 'tmp.mol2'), 'r') as f: view.addModel(f.read(), 'mol2', keepH=True) view.setStyle({'stick': {'radius': 0.2, 'color':'grey'}, 'sphere': {'scale': 0.3, 'colorscheme':modified_color_scheme}}) view.zoomTo() return view def _visualize_nglview(self, show_ports=False, color_scheme={}): """Visualize the Compound using nglview. Allows for visualization of a Compound within a Jupyter Notebook. Parameters ---------- show_ports : bool, optional, default=False Visualize Ports in addition to Particles """ nglview = import_('nglview') from mdtraj.geometry.sasa import _ATOMIC_RADII remove_digits = lambda x: ''.join(i for i in x if not i.isdigit() or i == '_') for particle in self.particles(): particle.name = remove_digits(particle.name).upper() if not particle.name: particle.name = 'UNK' tmp_dir = tempfile.mkdtemp() self.save(os.path.join(tmp_dir, 'tmp.mol2'), show_ports=show_ports, overwrite=True) widget = nglview.show_file(os.path.join(tmp_dir, 'tmp.mol2')) widget.clear() widget.add_ball_and_stick(cylinderOnly=True) elements = set([particle.name for particle in self.particles()]) scale = 50.0 for element in elements: try: widget.add_ball_and_stick('_{}'.format( element.upper()), aspect_ratio=_ATOMIC_RADII[element.title()]**1.5 * scale) except KeyError: ids = [str(i) for i, particle in enumerate(self.particles()) if particle.name == element] widget.add_ball_and_stick( '@{}'.format( ','.join(ids)), aspect_ratio=0.17**1.5 * scale, color='grey') if show_ports: widget.add_ball_and_stick('_VS', aspect_ratio=1.0, color='#991f00') overwrite_nglview_default(widget) return widget
[docs] def update_coordinates(self, filename, update_port_locations=True): """Update the coordinates of this Compound from a file. Parameters ---------- filename : str Name of file from which to load coordinates. Supported file types are the same as those supported by load() update_port_locations : bool, optional, default=True Update the locations of Ports so that they are shifted along with their anchor particles. Note: This conserves the location of Ports with respect to the anchor Particle, but does not conserve the orientation of Ports with respect to the molecule as a whole. See Also -------- load : Load coordinates from a file """ if update_port_locations: xyz_init = self.xyz self = load(filename, compound=self, coords_only=True) self._update_port_locations(xyz_init) else: self = load(filename, compound=self, coords_only=True)
def _update_port_locations(self, initial_coordinates): """Adjust port locations after particles have moved Compares the locations of Particles between 'self' and an array of reference coordinates. Shifts Ports in accordance with how far anchors have been moved. This conserves the location of Ports with respect to their anchor Particles, but does not conserve the orientation of Ports with respect to the molecule as a whole. Parameters ---------- initial_coordinates : np.ndarray, shape=(n, 3), dtype=float Reference coordinates to use for comparing how far anchor Particles have shifted. """ particles = list(self.particles()) for port in self.all_ports(): if port.anchor: idx = particles.index(port.anchor) shift = particles[idx].pos - initial_coordinates[idx] port.translate(shift) def _kick(self): """Slightly adjust all coordinates in a Compound Provides a slight adjustment to coordinates to kick them out of local energy minima. """ xyz_init = self.xyz for particle in self.particles(): particle.pos += (np.random.rand(3,) - 0.5) / 100 self._update_port_locations(xyz_init) warning_message = 'Please use Compound.energy_minimize()' @deprecated(warning_message) def energy_minimization(self, forcefield='UFF', steps=1000, **kwargs): self.energy_minimize(forcefield=forcefield, steps=steps, **kwargs)
[docs] def energy_minimize(self, forcefield='UFF', steps=1000, **kwargs): """Perform an energy minimization on a Compound Default behavior utilizes Open Babel (http://openbabel.org/docs/dev/) to perform an energy minimization/geometry optimization on a Compound by applying a generic force field Can also utilize OpenMM (http://openmm.org/) to energy minimize after atomtyping a Compound using Foyer (https://github.com/mosdef-hub/foyer) to apply a forcefield XML file that contains valid SMARTS strings. This function is primarily intended to be used on smaller components, with sizes on the order of 10's to 100's of particles, as the energy minimization scales poorly with the number of particles. Parameters ---------- steps : int, optional, default=1000 The number of optimization iterations forcefield : str, optional, default='UFF' The generic force field to apply to the Compound for minimization. Valid options are 'MMFF94', 'MMFF94s', ''UFF', 'GAFF', and 'Ghemical'. Please refer to the Open Babel documentation (http://open-babel. readthedocs.io/en/latest/Forcefields/Overview.html) when considering your choice of force field. Utilizing OpenMM for energy minimization requires a forcefield XML file with valid SMARTS strings. Please refer to (http://docs. openmm.org/7.0.0/userguide/application.html#creating-force-fields) for more information. Keyword Arguments ------------ algorithm : str, optional, default='cg' The energy minimization algorithm. Valid options are 'steep', 'cg', and 'md', corresponding to steepest descent, conjugate gradient, and equilibrium molecular dynamics respectively. For _energy_minimize_openbabel scale_bonds : float, optional, default=1 Scales the bond force constant (1 is completely on). For _energy_minimize_openmm scale_angles : float, optional, default=1 Scales the angle force constant (1 is completely on) For _energy_minimize_openmm scale_torsions : float, optional, default=1 Scales the torsional force constants (1 is completely on) For _energy_minimize_openmm Note: Only Ryckaert-Bellemans style torsions are currently supported scale_nonbonded : float, optional, default=1 Scales epsilon (1 is completely on) For _energy_minimize_openmm References ---------- If using _energy_minimize_openmm(), please cite: .. [1] P. Eastman, M. S. Friedrichs, J. D. Chodera, R. J. Radmer, C. M. Bruns, J. P. Ku, K. A. Beauchamp, T. J. Lane, L.-P. Wang, D. Shukla, T. Tye, M. Houston, T. Stich, C. Klein, M. R. Shirts, and V. S. Pande. "OpenMM 4: A Reusable, Extensible, Hardware Independent Library for High Performance Molecular Simulation." J. Chem. Theor. Comput. 9(1): 461-469. (2013). If using _energy_minimize_openbabel(), please cite: .. [1] O'Boyle, N.M.; Banck, M.; James, C.A.; Morley, C.; Vandermeersch, T.; Hutchison, G.R. "Open Babel: An open chemical toolbox." (2011) J. Cheminf. 3, 33 .. [2] Open Babel, version X.X.X http://openbabel.org, (installed Month Year) If using the 'MMFF94' force field please also cite the following: .. [3] T.A. Halgren, "Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94." (1996) J. Comput. Chem. 17, 490-519 .. [4] T.A. Halgren, "Merck molecular force field. II. MMFF94 van der Waals and electrostatic parameters for intermolecular interactions." (1996) J. Comput. Chem. 17, 520-552 .. [5] T.A. Halgren, "Merck molecular force field. III. Molecular geometries and vibrational frequencies for MMFF94." (1996) J. Comput. Chem. 17, 553-586 .. [6] T.A. Halgren and R.B. Nachbar, "Merck molecular force field. IV. Conformational energies and geometries for MMFF94." (1996) J. Comput. Chem. 17, 587-615 .. [7] T.A. Halgren, "Merck molecular force field. V. Extension of MMFF94 using experimental data, additional computational data, and empirical rules." (1996) J. Comput. Chem. 17, 616-641 If using the 'MMFF94s' force field please cite the above along with: .. [8] T.A. Halgren, "MMFF VI. MMFF94s option for energy minimization studies." (1999) J. Comput. Chem. 20, 720-729 If using the 'UFF' force field please cite the following: .. [3] Rappe, A.K., Casewit, C.J., Colwell, K.S., Goddard, W.A. III, Skiff, W.M. "UFF, a full periodic table force field for molecular mechanics and molecular dynamics simulations." (1992) J. Am. Chem. Soc. 114, 10024-10039 If using the 'GAFF' force field please cite the following: .. [3] Wang, J., Wolf, R.M., Caldwell, J.W., Kollman, P.A., Case, D.A. "Development and testing of a general AMBER force field" (2004) J. Comput. Chem. 25, 1157-1174 If using the 'Ghemical' force field please cite the following: .. [3] T. Hassinen and M. Perakyla, "New energy terms for reduced protein models implemented in an off-lattice force field" (2001) J. Comput. Chem. 22, 1229-1242 """ tmp_dir = tempfile.mkdtemp() original = clone(self) self._kick() self.save(os.path.join(tmp_dir, 'un-minimized.mol2')) extension = os.path.splitext(forcefield)[-1] openbabel_ffs = ['MMFF94', 'MMFF94s', 'UFF', 'GAFF', 'Ghemical'] if forcefield in openbabel_ffs: self._energy_minimize_openbabel(tmp_dir, forcefield=forcefield, steps=steps, **kwargs) elif extension == '.xml': self._energy_minimize_openmm(tmp_dir, forcefield_files=forcefield, forcefield_name=None, steps=steps, **kwargs) else: self._energy_minimize_openmm(tmp_dir, forcefield_files=None, forcefield_name=forcefield, steps=steps, **kwargs) self.update_coordinates(os.path.join(tmp_dir, 'minimized.pdb'))
def _energy_minimize_openmm( self, tmp_dir, forcefield_files=None, forcefield_name=None, steps=1000, scale_bonds=1, scale_angles=1, scale_torsions=1, scale_nonbonded=1): """ Perform energy minimization using OpenMM Converts an mBuild Compound to a ParmEd Structure, applies a forcefield using Foyer, and creates an OpenMM System. Parameters ---------- forcefield_files : str or list of str, optional, default=None Forcefield files to load forcefield_name : str, optional, default=None Apply a named forcefield to the output file using the `foyer` package, e.g. 'oplsaa'. Forcefields listed here: https://github.com/mosdef-hub/foyer/tree/master/foyer/forcefields steps : int, optional, default=1000 Number of energy minimization iterations scale_bonds : float, optional, default=1 Scales the bond force constant (1 is completely on) scale_angles : float, optiona, default=1 Scales the angle force constant (1 is completely on) scale_torsions : float, optional, default=1 Scales the torsional force constants (1 is completely on) scale_nonbonded : float, optional, default=1 Scales epsilon (1 is completely on) Notes ----- Assumes a particular organization for the force groups (HarmonicBondForce, HarmonicAngleForce, RBTorsionForce, NonBondedForce) References ---------- .. [1] P. Eastman, M. S. Friedrichs, J. D. Chodera, R. J. Radmer, C. M. Bruns, J. P. Ku, K. A. Beauchamp, T. J. Lane, L.-P. Wang, D. Shukla, T. Tye, M. Houston, T. Stich, C. Klein, M. R. Shirts, and V. S. Pande. "OpenMM 4: A Reusable, Extensible, Hardware Independent Library for High Performance Molecular Simulation." J. Chem. Theor. Comput. 9(1): 461-469. (2013). """ foyer = import_('foyer') to_parmed = self.to_parmed() ff = foyer.Forcefield(forcefield_files=forcefield_files, name=forcefield_name) to_parmed = ff.apply(to_parmed) from simtk.openmm.app.simulation import Simulation from simtk.openmm.app.pdbreporter import PDBReporter from simtk.openmm.openmm import LangevinIntegrator import simtk.unit as u system = to_parmed.createSystem() # Create an OpenMM System # Create a Langenvin Integrator in OpenMM integrator = LangevinIntegrator(298 * u.kelvin, 1 / u.picosecond, 0.002 * u.picoseconds) # Create Simulation object in OpenMM simulation = Simulation(to_parmed.topology, system, integrator) # Loop through forces in OpenMM System and set parameters for force in system.getForces(): if type(force).__name__ == "HarmonicBondForce": for bond_index in range(force.getNumBonds()): atom1, atom2, r0, k = force.getBondParameters(bond_index) force.setBondParameters(bond_index, atom1, atom2, r0, k * scale_bonds) force.updateParametersInContext(simulation.context) elif type(force).__name__ == "HarmonicAngleForce": for angle_index in range(force.getNumAngles()): atom1, atom2, atom3, r0, k = force.getAngleParameters( angle_index) force.setAngleParameters(angle_index, atom1, atom2, atom3, r0, k * scale_angles) force.updateParametersInContext(simulation.context) elif type(force).__name__ == "RBTorsionForce": for torsion_index in range(force.getNumTorsions()): atom1, atom2, atom3, atom4, c0, c1, c2, c3, c4, c5 = force.getTorsionParameters( torsion_index) force.setTorsionParameters( torsion_index, atom1, atom2, atom3, atom4, c0 * scale_torsions, c1 * scale_torsions, c2 * scale_torsions, c3 * scale_torsions, c4 * scale_torsions, c5 * scale_torsions) force.updateParametersInContext(simulation.context) elif type(force).__name__ == "NonbondedForce": for nb_index in range(force.getNumParticles()): charge, sigma, epsilon = force.getParticleParameters( nb_index) force.setParticleParameters(nb_index, charge, sigma, epsilon * scale_nonbonded) force.updateParametersInContext(simulation.context) elif type(force).__name__ == "CMMotionRemover": pass else: warn( 'OpenMM Force {} is ' 'not currently supported in _energy_minimize_openmm. ' 'This Force will not be updated!'.format( type(force).__name__)) simulation.context.setPositions(to_parmed.positions) # Run energy minimization through OpenMM simulation.minimizeEnergy(maxIterations=steps) reporter = PDBReporter(os.path.join(tmp_dir, 'minimized.pdb'), 1) reporter.report( simulation, simulation.context.getState( getPositions=True)) def _energy_minimize_openbabel(self, tmp_dir, steps=1000, algorithm='cg', forcefield='UFF'): """Perform an energy minimization on a Compound Utilizes Open Babel (http://openbabel.org/docs/dev/) to perform an energy minimization/geometry optimization on a Compound by applying a generic force field. This function is primarily intended to be used on smaller components, with sizes on the order of 10's to 100's of particles, as the energy minimization scales poorly with the number of particles. Parameters ---------- steps : int, optionl, default=1000 The number of optimization iterations algorithm : str, optional, default='cg' The energy minimization algorithm. Valid options are 'steep', 'cg', and 'md', corresponding to steepest descent, conjugate gradient, and equilibrium molecular dynamics respectively. forcefield : str, optional, default='UFF' The generic force field to apply to the Compound for minimization. Valid options are 'MMFF94', 'MMFF94s', ''UFF', 'GAFF', and 'Ghemical'. Please refer to the Open Babel documentation (http://open-babel. readthedocs.io/en/latest/Forcefields/Overview.html) when considering your choice of force field. References ---------- .. [1] O'Boyle, N.M.; Banck, M.; James, C.A.; Morley, C.; Vandermeersch, T.; Hutchison, G.R. "Open Babel: An open chemical toolbox." (2011) J. Cheminf. 3, 33 .. [2] Open Babel, version X.X.X http://openbabel.org, (installed Month Year) If using the 'MMFF94' force field please also cite the following: .. [3] T.A. Halgren, "Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94." (1996) J. Comput. Chem. 17, 490-519 .. [4] T.A. Halgren, "Merck molecular force field. II. MMFF94 van der Waals and electrostatic parameters for intermolecular interactions." (1996) J. Comput. Chem. 17, 520-552 .. [5] T.A. Halgren, "Merck molecular force field. III. Molecular geometries and vibrational frequencies for MMFF94." (1996) J. Comput. Chem. 17, 553-586 .. [6] T.A. Halgren and R.B. Nachbar, "Merck molecular force field. IV. Conformational energies and geometries for MMFF94." (1996) J. Comput. Chem. 17, 587-615 .. [7] T.A. Halgren, "Merck molecular force field. V. Extension of MMFF94 using experimental data, additional computational data, and empirical rules." (1996) J. Comput. Chem. 17, 616-641 If using the 'MMFF94s' force field please cite the above along with: .. [8] T.A. Halgren, "MMFF VI. MMFF94s option for energy minimization studies." (1999) J. Comput. Chem. 20, 720-729 If using the 'UFF' force field please cite the following: .. [3] Rappe, A.K., Casewit, C.J., Colwell, K.S., Goddard, W.A. III, Skiff, W.M. "UFF, a full periodic table force field for molecular mechanics and molecular dynamics simulations." (1992) J. Am. Chem. Soc. 114, 10024-10039 If using the 'GAFF' force field please cite the following: .. [3] Wang, J., Wolf, R.M., Caldwell, J.W., Kollman, P.A., Case, D.A. "Development and testing of a general AMBER force field" (2004) J. Comput. Chem. 25, 1157-1174 If using the 'Ghemical' force field please cite the following: .. [3] T. Hassinen and M. Perakyla, "New energy terms for reduced protein models implemented in an off-lattice force field" (2001) J. Comput. Chem. 22, 1229-1242 """ openbabel = import_('openbabel') for particle in self.particles(): try: get_by_symbol(particle.name) except KeyError: raise MBuildError("Element name {} not recognized. Cannot " "perform minimization." "".format(particle.name)) obConversion = openbabel.OBConversion() obConversion.SetInAndOutFormats("mol2", "pdb") mol = openbabel.OBMol() obConversion.ReadFile(mol, os.path.join(tmp_dir, "un-minimized.mol2")) ff = openbabel.OBForceField.FindForceField(forcefield) if ff is None: raise MBuildError("Force field '{}' not supported for energy " "minimization. Valid force fields are 'MMFF94', " "'MMFF94s', 'UFF', 'GAFF', and 'Ghemical'." "".format(forcefield)) warn( "Performing energy minimization using the Open Babel package. Please " "refer to the documentation to find the appropriate citations for " "Open Babel and the {} force field".format(forcefield)) ff.Setup(mol) if algorithm == 'steep': ff.SteepestDescent(steps) elif algorithm == 'md': ff.MolecularDynamicsTakeNSteps(steps, 300) elif algorithm == 'cg': ff.ConjugateGradients(steps) else: raise MBuildError("Invalid minimization algorithm. Valid options " "are 'steep', 'cg', and 'md'.") ff.UpdateCoordinates(mol) obConversion.WriteFile(mol, os.path.join(tmp_dir, 'minimized.pdb'))
[docs] def save(self, filename, show_ports=False, forcefield_name=None, forcefield_files=None, forcefield_debug=False, box=None, overwrite=False, residues=None, combining_rule='lorentz', foyer_kwargs=None, **kwargs): """Save the Compound to a file. Parameters ---------- filename : str Filesystem path in which to save the trajectory. The extension or prefix will be parsed and control the format. Supported extensions are: 'hoomdxml', 'gsd', 'gro', 'top', 'lammps', 'lmp', 'mcf' show_ports : bool, optional, default=False Save ports contained within the compound. forcefield_files : str, optional, default=None Apply a forcefield to the output file using a forcefield provided by the `foyer` package. forcefield_name : str, optional, default=None Apply a named forcefield to the output file using the `foyer` package, e.g. 'oplsaa'. Forcefields listed here: https://github.com/mosdef-hub/foyer/tree/master/foyer/forcefields forcefield_debug : bool, optional, default=False Choose level of verbosity when applying a forcefield through `foyer`. Specifically, when missing atom types in the forcefield xml file, determine if the warning is condensed or verbose. box : mb.Box, optional, default=self.boundingbox (with buffer) Box information to be written to the output file. If 'None', a bounding box is used with 0.25nm buffers at each face to avoid overlapping atoms. overwrite : bool, optional, default=False Overwrite if the filename already exists residues : str of list of str Labels of residues in the Compound. Residues are assigned by checking against Compound.name. combining_rule : str, optional, default='lorentz' Specify the combining rule for nonbonded interactions. Only relevant when the `foyer` package is used to apply a forcefield. Valid options are 'lorentz' and 'geometric', specifying Lorentz-Berthelot and geometric combining rules respectively. foyer_kwargs : dict, optional, default=None Keyword arguments to provide to `foyer.Forcefield.apply`. **kwargs Depending on the file extension these will be passed to either `write_gsd`, `write_hoomdxml`, `write_lammpsdata`, `write_mcf`, or `parmed.Structure.save`. See https://parmed.github.io/ParmEd/html/structobj/parmed.structure.Structure.html#parmed.structure.Structure.save Other Parameters ---------------- ref_distance : float, optional, default=1.0 Normalization factor used when saving to .gsd and .hoomdxml formats for converting distance values to reduced units. ref_energy : float, optional, default=1.0 Normalization factor used when saving to .gsd and .hoomdxml formats for converting energy values to reduced units. ref_mass : float, optional, default=1.0 Normalization factor used when saving to .gsd and .hoomdxml formats for converting mass values to reduced units. atom_style: str, default='full' Defines the style of atoms to be saved in a LAMMPS data file. The following atom styles are currently supported: 'full', 'atomic', 'charge', 'molecular' see http://lammps.sandia.gov/doc/atom_style.html for more information on atom styles. unit_style: str, default='real' Defines to unit style to be save in a LAMMPS data file. Defaults to 'real' units. Current styles are supported: 'real', 'lj' see https://lammps.sandia.gov/doc/99/units.html for more information on unit styles Notes ------ When saving the compound as a json, only the following arguments are used: - filename - show_ports See Also -------- formats.gsdwrite.write_gsd : Write to GSD format formats.hoomdxml.write_hoomdxml : Write to Hoomd XML format formats.xyzwriter.write_xyz : Write to XYZ format formats.lammpsdata.write_lammpsdata : Write to LAMMPS data format formats.cassandramcf.write_mcf : Write to Cassandra MCF format formats.json_formats.compound_to_json : Write to a json file """ extension = os.path.splitext(filename)[-1] if extension == '.json': compound_to_json(self, file_path=filename, include_ports=show_ports) return # Savers supported by mbuild.formats savers = {'.hoomdxml': write_hoomdxml, '.gsd': write_gsd, '.xyz': write_xyz, '.lammps': write_lammpsdata, '.lmp': write_lammpsdata, '.par': write_par,} if has_networkx: from mbuild.formats.cassandramcf import write_mcf savers.update({'.mcf': write_mcf}) try: saver = savers[extension] except KeyError: saver = None if os.path.exists(filename) and not overwrite: raise IOError('{0} exists; not overwriting'.format(filename)) structure = self.to_parmed(box=box, residues=residues, show_ports=show_ports) # Apply a force field with foyer if specified if forcefield_name or forcefield_files: foyer = import_('foyer') ff = foyer.Forcefield(forcefield_files=forcefield_files, name=forcefield_name, debug=forcefield_debug) if not foyer_kwargs: foyer_kwargs = {} structure = ff.apply(structure, **foyer_kwargs) structure.combining_rule = combining_rule total_charge = sum([atom.charge for atom in structure]) if round(total_charge, 4) != 0.0: warn('System is not charge neutral. Total charge is {}.' ''.format(total_charge)) # Provide a warning if rigid_ids are not sequential from 0 if self.contains_rigid: unique_rigid_ids = sorted(set([ p.rigid_id for p in self.rigid_particles()])) if max(unique_rigid_ids) != len(unique_rigid_ids) - 1: warn("Unique rigid body IDs are not sequential starting from zero.") if saver: # mBuild supported saver. if extension in ['.gsd', '.hoomdxml']: kwargs['rigid_bodies'] = [ p.rigid_id for p in self.particles()] # lammps does not require the box to be centered at any a specific origin # min and max dimensions are therefore needed to write the file in a consistent way # the parmed structure only has the box length if extension in ['.lammps', '.lmp']: if box: kwargs['mins'] = [m for m in box.mins] kwargs['maxs'] = [m for m in box.maxs] saver(filename=filename, structure=structure, **kwargs) elif extension == '.sdf': pybel = import_('pybel') new_compound = Compound() # Convert pmd.Structure to mb.Compound new_compound.from_parmed(structure) # Convert mb.Compound to pybel molecule pybel_molecule = new_compound.to_pybel() # Write out pybel molecule to SDF file output_sdf = pybel.Outputfile("sdf", filename, overwrite=overwrite) output_sdf.write(pybel_molecule) output_sdf.close() else: # ParmEd supported saver. structure.save(filename, overwrite=overwrite, **kwargs)
[docs] def translate(self, by): """Translate the Compound by a vector Parameters ---------- by : np.ndarray, shape=(3,), dtype=float """ new_positions = _translate(self.xyz_with_ports, by) self.xyz_with_ports = new_positions
[docs] def translate_to(self, pos): """Translate the Compound to a specific position Parameters ---------- pos : np.ndarray, shape=3(,), dtype=float """ self.translate(pos - self.center)
[docs] def rotate(self, theta, around): """Rotate Compound around an arbitrary vector. Parameters ---------- theta : float The angle by which to rotate the Compound, in radians. around : np.ndarray, shape=(3,), dtype=float The vector about which to rotate the Compound. """ new_positions = _rotate(self.xyz_with_ports, theta, around) self.xyz_with_ports = new_positions
[docs] def spin(self, theta, around): """Rotate Compound in place around an arbitrary vector. Parameters ---------- theta : float The angle by which to rotate the Compound, in radians. around : np.ndarray, shape=(3,), dtype=float The axis about which to spin the Compound. """ around = np.asarray(around).reshape(3) center_pos = self.center self.translate(-center_pos) self.rotate(theta, around) self.translate(center_pos)
# Interface to Trajectory for reading/writing .pdb and .mol2 files. # -----------------------------------------------------------------
[docs] def from_trajectory(self, traj, frame=-1, coords_only=False, infer_hierarchy=True): """Extract atoms and bonds from a md.Trajectory. Will create sub-compounds for every chain if there is more than one and sub-sub-compounds for every residue. Parameters ---------- traj : mdtraj.Trajectory The trajectory to load. frame : int, optional, default=-1 (last) The frame to take coordinates from. coords_only : bool, optional, default=False Only read coordinate information infer_hierarchy : bool, optional, default=True If True, infer compound hierarchy from chains and residues """ if coords_only: if traj.n_atoms != self.n_particles: raise ValueError('Number of atoms in {traj} does not match' ' {self}'.format(**locals())) atoms_particles = zip(traj.topology.atoms, self.particles(include_ports=False)) if None in self._particles(include_ports=False): raise ValueError('Some particles are None') for mdtraj_atom, particle in atoms_particles: particle.pos = traj.xyz[frame, mdtraj_atom.index] return atom_mapping = dict() for chain in traj.topology.chains: if traj.topology.n_chains > 1: chain_compound = Compound() self.add(chain_compound, 'chain[$]') else: chain_compound = self for res in chain.residues: if infer_hierarchy: res_compound = Compound(name=res.name) chain_compound.add(res_compound) parent_cmpd = res_compound else: parent_cmpd = chain_compound for atom in res.atoms: new_atom = Particle(name=str(atom.name), pos=traj.xyz[frame, atom.index]) parent_cmpd.add( new_atom, label='{0}[$]'.format( atom.name)) atom_mapping[atom] = new_atom for mdtraj_atom1, mdtraj_atom2 in traj.topology.bonds: atom1 = atom_mapping[mdtraj_atom1] atom2 = atom_mapping[mdtraj_atom2] self.add_bond((atom1, atom2)) if np.any(traj.unitcell_lengths) and np.any(traj.unitcell_lengths[0]): self.periodicity = traj.unitcell_lengths[0] else: self.periodicity = np.array([0., 0., 0.])
[docs] def to_trajectory(self, show_ports=False, chains=None, residues=None, box=None): """Convert to an md.Trajectory and flatten the compound. Parameters ---------- show_ports : bool, optional, default=False Include all port atoms when converting to trajectory. chains : mb.Compound or list of mb.Compound Chain types to add to the topology residues : str of list of str Labels of residues in the Compound. Residues are assigned by checking against Compound.name. box : mb.Box, optional, default=self.boundingbox (with buffer) Box information to be used when converting to a `Trajectory`. If 'None', a bounding box is used with a 0.5nm buffer in each dimension. to avoid overlapping atoms, unless `self.periodicity` is not None, in which case those values are used for the box lengths. Returns ------- trajectory : md.Trajectory See also -------- _to_topology """ atom_list = [particle for particle in self.particles(show_ports)] top = self._to_topology(atom_list, chains, residues) # Coordinates. xyz = np.ndarray(shape=(1, top.n_atoms, 3), dtype='float') for idx, atom in enumerate(atom_list): xyz[0, idx] = atom.pos # Unitcell information. unitcell_angles = [90.0, 90.0, 90.0] if box is None: unitcell_lengths = np.empty(3) for dim, val in enumerate(self.periodicity): if val: unitcell_lengths[dim] = val else: unitcell_lengths[dim] = self.boundingbox.lengths[dim] + 0.5 else: unitcell_lengths = box.lengths unitcell_angles = box.angles return md.Trajectory(xyz, top, unitcell_lengths=unitcell_lengths, unitcell_angles=unitcell_angles)
def _to_topology(self, atom_list, chains=None, residues=None): """Create a mdtraj.Topology from a Compound. Parameters ---------- atom_list : list of mb.Compound Atoms to include in the topology chains : mb.Compound or list of mb.Compound Chain types to add to the topology residues : str of list of str Labels of residues in the Compound. Residues are assigned by checking against Compound.name. Returns ------- top : mdtraj.Topology See Also -------- mdtraj.Topology : Details on the mdtraj Topology object """ from mdtraj.core.topology import Topology if isinstance(chains, str): chains = [chains] if isinstance(chains, (list, set)): chains = tuple(chains) if isinstance(residues, str): residues = [residues] if isinstance(residues, (list, set)): residues = tuple(residues) top = Topology() atom_mapping = {} default_chain = top.add_chain() default_residue = top.add_residue('RES', default_chain) compound_residue_map = dict() atom_residue_map = dict() compound_chain_map = dict() atom_chain_map = dict() for atom in atom_list: # Chains if chains: if atom.name in chains: current_chain = top.add_chain() compound_chain_map[atom] = current_chain else: for parent in atom.ancestors(): if chains and parent.name in chains: if parent not in compound_chain_map: current_chain = top.add_chain() compound_chain_map[parent] = current_chain current_residue = top.add_residue( 'RES', current_chain) break else: current_chain = default_chain else: current_chain = default_chain atom_chain_map[atom] = current_chain # Residues if residues: if atom.name in residues: current_residue = top.add_residue(atom.name, current_chain) compound_residue_map[atom] = current_residue else: for parent in atom.ancestors(): if residues and parent.name in residues: if parent not in compound_residue_map: current_residue = top.add_residue( parent.name, current_chain) compound_residue_map[parent] = current_residue break else: current_residue = default_residue else: if chains: try: # Grab the default residue from the custom chain. current_residue = next(current_chain.residues) except StopIteration: # Add the residue to the current chain current_residue = top.add_residue('RES', current_chain) else: # Grab the default chain's default residue current_residue = default_residue atom_residue_map[atom] = current_residue # Add the actual atoms try: elem = get_by_symbol(atom.name) except KeyError: elem = get_by_symbol("VS") at = top.add_atom(atom.name, elem, atom_residue_map[atom]) at.charge = atom.charge atom_mapping[atom] = at # Remove empty default residues. chains_to_remove = [ chain for chain in top.chains if chain.n_atoms == 0] residues_to_remove = [res for res in top.residues if res.n_atoms == 0] for chain in chains_to_remove: top._chains.remove(chain) for res in residues_to_remove: for chain in top.chains: try: chain._residues.remove(res) except ValueError: # Already gone. pass for atom1, atom2 in self.bonds(): # Ensure that both atoms are part of the compound. This becomes an # issue if you try to convert a sub-compound to a topology which is # bonded to a different subcompound. if all(a in atom_mapping.keys() for a in [atom1, atom2]): top.add_bond(atom_mapping[atom1], atom_mapping[atom2]) return top
[docs] def from_parmed(self, structure, coords_only=False, infer_hierarchy=True): """Extract atoms and bonds from a pmd.Structure. Will create sub-compounds for every chain if there is more than one and sub-sub-compounds for every residue. Parameters ---------- structure : pmd.Structure The structure to load. coords_only : bool Set preexisting atoms in compound to coordinates given by structure. infer_hierarchy : bool, optional, default=True If true, infer compound hierarchy from chains and residues """ if coords_only: if len(structure.atoms) != self.n_particles: raise ValueError( 'Number of atoms in {structure} does not match' ' {self}'.format( **locals())) atoms_particles = zip(structure.atoms, self.particles(include_ports=False)) if None in self._particles(include_ports=False): raise ValueError('Some particles are None') for parmed_atom, particle in atoms_particles: particle.pos = np.array([parmed_atom.xx, parmed_atom.xy, parmed_atom.xz]) / 10 return atom_mapping = dict() chain_id = None chains = defaultdict(list) for residue in structure.residues: chains[residue.chain].append(residue) for chain, residues in chains.items(): if len(chains) > 1: chain_compound = Compound() self.add(chain_compound, chain_id) else: chain_compound = self for residue in residues: if infer_hierarchy: residue_compound = Compound(name=residue.name) chain_compound.add(residue_compound) parent_cmpd = residue_compound else: parent_cmpd = chain_compound for atom in residue.atoms: pos = np.array([atom.xx, atom.xy, atom.xz]) / 10 new_atom = Particle(name=str(atom.name), pos=pos) parent_cmpd.add( new_atom, label='{0}[$]'.format( atom.name)) atom_mapping[atom] = new_atom for bond in structure.bonds: atom1 = atom_mapping[bond.atom1] atom2 = atom_mapping[bond.atom2] self.add_bond((atom1, atom2)) if structure.box is not None: # Convert from A to nm self.periodicity = 0.1 * structure.box[0:3] else: self.periodicity = np.array([0., 0., 0.])
[docs] def to_parmed(self, box=None, title='', residues=None, show_ports=False, infer_residues=False): """Create a ParmEd Structure from a Compound. Parameters ---------- box : mb.Box, optional, default=self.boundingbox (with buffer) Box information to be used when converting to a `Structure`. If 'None', a bounding box is used with 0.25nm buffers at each face to avoid overlapping atoms, unless `self.periodicity` is not None, in which case those values are used for the box lengths. title : str, optional, default=self.name Title/name of the ParmEd Structure residues : str of list of str Labels of residues in the Compound. Residues are assigned by checking against Compound.name. show_ports : boolean, optional, default=False Include all port atoms when converting to a `Structure`. infer_residues : bool, optional, default=False Attempt to assign residues based on names of children. Returns ------- parmed.structure.Structure ParmEd Structure object converted from self See Also -------- parmed.structure.Structure : Details on the ParmEd Structure object """ structure = pmd.Structure() structure.title = title if title else self.name atom_mapping = {} # For creating bonds below guessed_elements = set() # Attempt to grab residue names based on names of children if not residues and infer_residues: residues = list(set([child.name for child in self.children])) if isinstance(residues, str): residues = [residues] if isinstance(residues, (list, set)): residues = tuple(residues) default_residue = pmd.Residue('RES') port_residue = pmd.Residue('PRT') compound_residue_map = dict() atom_residue_map = dict() # Loop through particles and add initialize ParmEd atoms for atom in self.particles(include_ports=show_ports): if atom.port_particle: current_residue = port_residue atom_residue_map[atom] = current_residue if current_residue not in structure.residues: structure.residues.append(current_residue) pmd_atom = pmd.Atom(atomic_number=0, name='VS', mass=0, charge=0) pmd_atom.xx, pmd_atom.xy, pmd_atom.xz = atom.pos * 10 # Angstroms else: if residues and atom.name in residues: current_residue = pmd.Residue(atom.name) atom_residue_map[atom] = current_residue compound_residue_map[atom] = current_residue elif residues: for parent in atom.ancestors(): if residues and parent.name in residues: if parent not in compound_residue_map: current_residue = pmd.Residue(parent.name) compound_residue_map[parent] = current_residue atom_residue_map[atom] = current_residue break else: # Did not find specified residues in ancestors. current_residue = default_residue atom_residue_map[atom] = current_residue else: current_residue = default_residue atom_residue_map[atom] = current_residue if current_residue not in structure.residues: structure.residues.append(current_residue) atomic_number = None name = ''.join(char for char in atom.name if not char.isdigit()) try: atomic_number = AtomicNum[atom.name.capitalize()] except KeyError: element = element_by_name(atom.name.capitalize()) if name not in guessed_elements: warn( 'Guessing that "{}" is element: "{}"'.format( atom, element)) guessed_elements.add(name) else: element = atom.name.capitalize() atomic_number = atomic_number or AtomicNum[element] mass = Mass[element] pmd_atom = pmd.Atom(atomic_number=atomic_number, name=atom.name, mass=mass, charge=atom.charge) pmd_atom.xx, pmd_atom.xy, pmd_atom.xz = atom.pos * 10 # Angstroms residue = atom_residue_map[atom] structure.add_atom(pmd_atom, resname=residue.name, resnum=residue.idx) atom_mapping[atom] = pmd_atom # "Claim" all of the items it contains and subsequently index all of its items structure.residues.claim() # Create and add bonds to ParmEd Structure for atom1, atom2 in self.bonds(): bond = pmd.Bond(atom_mapping[atom1], atom_mapping[atom2]) structure.bonds.append(bond) # pad box with .25nm buffers if box is None: box = self.boundingbox box_vec_max = box.maxs.tolist() box_vec_min = box.mins.tolist() for dim, val in enumerate(self.periodicity): if val: box_vec_max[dim] = val box_vec_min[dim] = 0.0 if not val: box_vec_max[dim] += 0.25 box_vec_min[dim] -= 0.25 box = Box(mins=box_vec_min, maxs=box_vec_max) box_vector = np.empty(6) if box.angles is not None: box_vector[3:6] = box.angles else: box_vector[3] = box_vector[4] = box_vector[5] = 90.0 for dim in range(3): box_vector[dim] = box.lengths[dim] * 10 structure.box = box_vector return structure
[docs] def to_networkx(self, names_only=False): """Create a NetworkX graph representing the hierarchy of a Compound. Parameters ---------- names_only : bool, optional, default=False Store only the names of the compounds in the graph, appended with their IDs, for distinction even if they have the same name. When set to False, the default behavior, the nodes are the compounds themselves. Returns ------- G : networkx.DiGraph Notes ----- This digraph is not the bondgraph of the compound. See Also -------- mbuild.bond_graph """ nx = import_('networkx') nodes = list() edges = list() if names_only: nodes.append(self.name + '_' + str(id(self))) else: nodes.append(self) nodes, edges = self._iterate_children(nodes, edges, names_only=names_only) graph = nx.DiGraph() graph.add_nodes_from(nodes) graph.add_edges_from(edges) return graph
def _iterate_children(self, nodes, edges, names_only=False): """ Create nodes and edges that connect parents and their corresponding children""" if not self.children: return nodes, edges for child in self.children: if names_only: unique_name = child.name + '_' + str(id(child)) unique_name_parent = child.parent.name + '_' + str((id(child.parent))) nodes.append(unique_name) edges.append([unique_name_parent, unique_name]) else: nodes.append(child) edges.append([child.parent, child]) nodes, edges = child._iterate_children(nodes, edges, names_only=names_only) return nodes, edges
[docs] def to_pybel(self, box=None, title='', residues=None, show_ports=False, infer_residues=False): """ Create a pybel.Molecule from a Compound Parameters --------- box : mb.Box, def None title : str, optional, default=self.name Title/name of the ParmEd Structure residues : str of list of str Labels of residues in the Compound. Residues are assigned by checking against Compound.name. show_ports : boolean, optional, default=False Include all port atoms when converting to a `Structure`. infer_residues : bool, optional, default=False Attempt to assign residues based on names of children Returns ------ pybel.Molecule Notes ----- Most of the mb.Compound is first converted to openbabel.OBMol And then pybel creates a pybel.Molecule from the OBMol Bond orders are assumed to be 1 OBMol atom indexing starts at 1, with spatial dimension Angstrom """ openbabel = import_('openbabel') pybel = import_('pybel') mol = openbabel.OBMol() particle_to_atom_index = {} if not residues and infer_residues: residues = list(set([child.name for child in self.children])) if isinstance(residues, str): residues = [residues] if isinstance(residues, (list, set)): residues = tuple(residues) compound_residue_map = dict() atom_residue_map = dict() for i, part in enumerate(self.particles(include_ports=show_ports)): if residues and part.name in residues: current_residue = mol.NewResidue() current_residue.SetName(part.name) atom_residue_map[part] = current_residue compound_residue_map[part] = current_residue elif residues: for parent in part.ancestors(): if residues and parent.name in residues: if parent not in compound_residue_map: current_residue = mol.NewResidue() current_residue.SetName(parent.name) compound_residue_map[parent] = current_residue atom_residue_map[part] = current_residue break else: # Did not find specified residues in ancestors. current_residue = mol.NewResidue() current_residue.SetName("RES") atom_residue_map[part] = current_residue else: current_residue = mol.NewResidue() current_residue.SetName("RES") atom_residue_map[part] = current_residue temp = mol.NewAtom() residue = atom_residue_map[part] temp.SetResidue(residue) if part.port_particle: temp.SetAtomicNum(0) else: try: temp.SetAtomicNum(AtomicNum[part.name.capitalize()]) except KeyError: warn("Could not infer atomic number from " "{}, setting to 0".format(part.name)) temp.SetAtomicNum(0) temp.SetVector(*(part.xyz[0]*10)) particle_to_atom_index[part] = i ucell = openbabel.OBUnitCell() if box is None: box = self.boundingbox a, b, c = 10.0 * box.lengths alpha, beta, gamma = np.radians(box.angles) cosa = np.cos(alpha) cosb = np.cos(beta) sinb = np.sin(beta) cosg = np.cos(gamma) sing = np.sin(gamma) mat_coef_y = (cosa - cosb * cosg) / sing mat_coef_z = np.power(sinb, 2, dtype=float) - \ np.power(mat_coef_y, 2, dtype=float) if mat_coef_z > 0.: mat_coef_z = np.sqrt(mat_coef_z) else: raise Warning('Non-positive z-vector. Angles {} ' 'do not generate a box with the z-vector in the' 'positive z direction'.format(box.angles)) box_vec = [[1, 0, 0], [cosg, sing, 0], [cosb, mat_coef_y, mat_coef_z]] box_vec = np.asarray(box_vec) box_mat = (np.array([a,b,c])* box_vec.T).T first_vector = openbabel.vector3(*box_mat[0]) second_vector = openbabel.vector3(*box_mat[1]) third_vector = openbabel.vector3(*box_mat[2]) ucell.SetData(first_vector, second_vector, third_vector) mol.CloneData(ucell) for bond in self.bonds(): bond_order = 1 mol.AddBond(particle_to_atom_index[bond[0]]+1, particle_to_atom_index[bond[1]]+1, bond_order) pybelmol = pybel.Molecule(mol) pybelmol.title = title if title else self.name return pybelmol
[docs] def from_pybel(self, pybel_mol, use_element=True, coords_only=False, infer_hierarchy=True, ignore_box_warn=False): """Create a Compound from a Pybel.Molecule Parameters --------- pybel_mol: pybel.Molecule use_element : bool, default True If True, construct mb Particles based on the pybel Atom's element. If False, construcs mb Particles based on the pybel Atom's type coords_only : bool, default False Set preexisting atoms in compound to coordinates given by structure. Note: Not yet implemented, included only for parity with other conversion functions infer_hierarchy : bool, optional, default=True If True, infer hierarchy from residues ignore_box_warn : bool, optional, default=False If True, ignore warning if no box is present. """ openbabel = import_("openbabel") self.name = pybel_mol.title.split('.')[0] resindex_to_cmpd = {} if coords_only: raise Warning('coords_only=True not yet implemented for ' 'conversion from pybel') # Iterating through pybel_mol for atom/residue information # This could just as easily be implemented by # an OBMolAtomIter from the openbabel library, # but this seemed more convenient at time of writing # pybel atoms are 1-indexed, coordinates in Angstrom for atom in pybel_mol.atoms: xyz = np.array(atom.coords)/10 if use_element: try: temp_name = Element[atom.atomicnum] except KeyError: warn("No element detected for atom at index " "{} with number {}, type {}".format( atom.idx, atom.atomicnum, atom.type)) temp_name = atom.type else: temp_name = atom.type temp = Particle(name=temp_name, pos=xyz) if infer_hierarchy and hasattr(atom, 'residue'): # Is there a safer way to check for res? if atom.residue.idx not in resindex_to_cmpd: res_cmpd = Compound(name=atom.residue.name) resindex_to_cmpd[atom.residue.idx] = res_cmpd self.add(res_cmpd) resindex_to_cmpd[atom.residue.idx].add(temp) else: self.add(temp) # Iterating through pybel_mol.OBMol for bond information # Bonds are 0-indexed, but the atoms are 1-indexed # Bond information doesn't appear stored in pybel_mol, # so we need to look into the OBMol object, # using an iterator from the openbabel library for bond in openbabel.OBMolBondIter(pybel_mol.OBMol): self.add_bond([self[bond.GetBeginAtomIdx()-1], self[bond.GetEndAtomIdx()-1]]) if hasattr(pybel_mol, 'unitcell'): box = Box(lengths=[pybel_mol.unitcell.GetA()/10, pybel_mol.unitcell.GetB()/10, pybel_mol.unitcell.GetC()/10], angles=[pybel_mol.unitcell.GetAlpha(), pybel_mol.unitcell.GetBeta(), pybel_mol.unitcell.GetGamma()]) self.periodicity = box.lengths else: if not ignore_box_warn: warn("No unitcell detected for pybel.Molecule {}".format(pybel_mol))
# TODO: Decide how to gather PBC information from openbabel. Options may # include storing it in .periodicity or writing a separate function # that returns the box.
[docs] def to_intermol(self, molecule_types=None): # pragma: no cover """Create an InterMol system from a Compound. Parameters ---------- molecule_types : list or tuple of subclasses of Compound Returns ------- intermol_system : intermol.system.System """ from intermol.atom import Atom as InterMolAtom from intermol.molecule import Molecule from intermol.system import System import simtk.unit as u if isinstance(molecule_types, list): molecule_types = tuple(molecule_types) elif molecule_types is None: molecule_types = (type(self),) intermol_system = System() last_molecule_compound = None for atom_index, atom in enumerate(self.particles()): for parent in atom.ancestors(): # Don't want inheritance via isinstance(). if type(parent) in molecule_types: # Check if we have encountered this molecule type before. if parent.name not in intermol_system.molecule_types: self._add_intermol_molecule_type( intermol_system, parent) if parent != last_molecule_compound: last_molecule_compound = parent last_molecule = Molecule(name=parent.name) intermol_system.add_molecule(last_molecule) break else: # Should never happen if molecule_types only contains # type(self) raise ValueError( 'Found an atom {} that is not part of any of ' 'the specified molecule types {}'.format( atom, molecule_types)) # Add the actual intermol atoms. intermol_atom = InterMolAtom(atom_index + 1, name=atom.name, residue_index=1, residue_name='RES') intermol_atom.position = atom.pos * u.nanometers last_molecule.add_atom(intermol_atom) return intermol_system
[docs] def get_smiles(self): """Get SMILES string for compound Bond order is guessed with pybel and may lead to incorrect SMILES strings. Returns ------- smiles_string: str """ pybel_cmp = self.to_pybel() pybel_cmp.OBMol.PerceiveBondOrders() # we only need the smiles string smiles = pybel_cmp.write().split()[0] return smiles
@staticmethod def _add_intermol_molecule_type(intermol_system, parent): # pragma: no cover """Create a molecule type for the parent and add bonds. This method takes an intermol system and adds a parent compound, including its particles and bonds, to it. """ from intermol.moleculetype import MoleculeType from intermol.forces.bond import Bond as InterMolBond molecule_type = MoleculeType(name=parent.name) intermol_system.add_molecule_type(molecule_type) for index, parent_atom in enumerate(parent.particles()): parent_atom.index = index + 1 for atom1, atom2 in parent.bonds(): intermol_bond = InterMolBond(atom1.index, atom2.index) molecule_type.bonds.add(intermol_bond) def __getitem__(self, selection): if isinstance(selection, int): return list(self.particles())[selection] if isinstance(selection, str): if selection not in self.labels: raise MBuildError('{}[\'{}\'] does not exist.'.format(self.name,selection)) return self.labels.get(selection) def __repr__(self): descr = list('<') descr.append(self.name + ' ') if self.children: descr.append('{:d} particles, '.format(self.n_particles)) if any(self.periodicity): descr.append('periodicity: {}, '.format(self.periodicity)) else: descr.append('non-periodic, ') else: descr.append('pos=({: .4f},{: .4f},{: .4f}), '.format(*self.pos)) descr.append('{:d} bonds, '.format(self.n_bonds)) descr.append('id: {}>'.format(id(self))) return ''.join(descr) def _clone(self, clone_of=None, root_container=None): """A faster alternative to deepcopying. Does not resolve circular dependencies. This should be safe provided you never try to add the top of a Compound hierarchy to a sub-Compound. Clones compound hierarchy only, not the bonds. """ if root_container is None: root_container = self if clone_of is None: clone_of = dict() # If this compound has already been cloned, return that. if self in clone_of: return clone_of[self] # Otherwise we make a new clone. cls = self.__class__ newone = cls.__new__(cls) # Remember that we're cloning the new one of self. clone_of[self] = newone newone.name = deepcopy(self.name) newone.periodicity = deepcopy(self.periodicity) newone._pos = deepcopy(self._pos) newone.port_particle = deepcopy(self.port_particle) newone.box = deepcopy(self.box) newone._check_if_contains_rigid_bodies = deepcopy( self._check_if_contains_rigid_bodies) newone._contains_rigid = deepcopy(self._contains_rigid) newone._rigid_id = deepcopy(self._rigid_id) newone._charge = deepcopy(self._charge) if hasattr(self, 'index'): newone.index = deepcopy(self.index) if self.children is None: newone.children = None else: newone.children = OrderedSet() # Parent should be None initially. newone.parent = None newone.labels = OrderedDict() newone.referrers = set() newone.bond_graph = None # Add children to clone. if self.children: for child in self.children: newchild = child._clone(clone_of, root_container) newone.children.add(newchild) newchild.parent = newone # Copy labels, except bonds with atoms outside the hierarchy. if self.labels: for label, compound in self.labels.items(): if not isinstance(compound, list): newone.labels[label] = compound._clone( clone_of, root_container) compound.referrers.add(clone_of[compound]) else: # compound is a list of compounds, so we create an empty # list, and add the clones of the original list elements. newone.labels[label] = [] for subpart in compound: newone.labels[label].append( subpart._clone(clone_of, root_container)) # Referrers must have been handled already, or the will # be handled return newone def _clone_bonds(self, clone_of=None): """While cloning, clone the bond of the source compound to clone compound""" newone = clone_of[self] for c1, c2 in self.bonds(): try: newone.add_bond((clone_of[c1], clone_of[c2])) except KeyError: raise MBuildError( "Cloning failed. Compound contains bonds to " "Particles outside of its containment hierarchy.")
Particle = Compound