def ethane(self):
mytop = Topology()
c1 = Atom(name='C1')
h11 = Atom(name='H11')
h12 = Atom(name='H12')
h13 = Atom(name='H13')
c2 = Atom(name='C2')
h21 = Atom(name='H21')
h22 = Atom(name='H22')
h23 = Atom(name='H23')
c1h11 = Bond(connection_members=[c1, h11])
c1h12 = Bond(connection_members=[c1, h12])
c1h13 = Bond(connection_members=[c1, h13])
c2h21 = Bond(connection_members=[c2, h21])
c2h22 = Bond(connection_members=[c2, h22])
c2h23 = Bond(connection_members=[c2, h23])
c1c2 = Bond(connection_members=[c1, c2])
mytop.add_connection(c1h11, update_types=False)
mytop.add_connection(c1h12, update_types=False)
mytop.add_connection(c1h13, update_types=False)
mytop.add_connection(c2h21, update_types=False)
mytop.add_connection(c2h22, update_types=False)
mytop.add_connection(c2h23, update_types=False)
mytop.add_connection(c1c2, update_types=False)
mytop.update_topology()
return mytop
def test_with_1000_atom_types(self):
top = Topology()
for i in range(1000):
site = Site()
atom_type = AtomType()
site.atom_type = atom_type
top.add_site(site, update_types=False)
top.update_topology()
assert len(top.atom_types) == 1
assert top.n_sites == 1000
def test_add_duplicate_connected_atom(self):
top = Topology()
atom1 = Atom(name="AtomA")
atom2 = Atom(name="AtomB")
bond = Bond(connection_members=[atom1, atom2])
bond_eq = Bond(connection_members=[atom1, atom2])
top.add_connection(bond)
top.add_connection(bond_eq)
top.update_topology()
assert top.n_connections == 1
def test_topology_atom_type_changes(self):
top = Topology()
for i in range(100):
site = Site(name='site{}'.format(i))
atom_type = AtomType(name='atom_type{}'.format(i % 10))
site.atom_type = atom_type
top.add_site(site, update_types=False)
top.update_topology()
assert len(top.atom_types) == 10
top.sites[0].atom_type.name = 'atom_type_changed'
assert id(top.sites[0].atom_type) == id(top.sites[10].atom_type)
assert top.sites[10].atom_type.name == 'atom_type_changed'
assert top.is_typed()
def test_top_update(self):
top = Topology()
top.update_topology()
assert top.n_sites == 0
assert len(top.atom_types) == 0
assert len(top.atom_type_expressions) == 0
assert top.n_connections == 0
assert len(top.connection_types) == 0
assert len(top.connection_type_expressions) == 0
atomtype = AtomType()
site1 = Site(name='site1', atom_type=atomtype)
top.add_site(site1)
site2 = Site(name='site2', atom_type=atomtype)
top.add_site(site2)
assert top.n_sites == 2
assert len(top.atom_types) == 1
assert len(top.atom_type_expressions) == 1
assert top.n_connections == 0
assert len(top.connection_types) == 0
assert len(top.connection_type_expressions) == 0
ctype = BondType()
connection_12 = Bond(connection_members=[site1, site2],
connection_type=ctype)
top.add_connection(connection_12)
assert top.n_sites == 2
assert len(top.atom_types) == 1
assert len(top.atom_type_expressions) == 1
assert top.n_connections == 1
assert len(top.connection_types) == 1
assert len(top.connection_type_expressions) == 1
site1.atom_type = AtomType(expression='sigma*epsilon')
assert top.n_sites == 2
assert len(top.atom_types) == 1
assert len(top.atom_type_expressions) == 1
assert top.n_connections == 1
assert len(top.connection_types) == 1
assert len(top.connection_type_expressions) == 1
top.update_atom_types()
assert top.n_sites == 2
assert len(top.atom_types) == 2
assert len(top.atom_type_expressions) == 2
assert top.n_connections == 1
assert len(top.connection_types) == 1
assert len(top.connection_type_expressions) == 1
def test_top_update(self):
top = Topology()
top.update_topology()
assert top.n_sites == 0
assert len(top.atom_types) == 0
assert len(top.atom_type_expressions) == 0
assert top.n_connections == 0
assert len(top.connection_types) == 0
assert len(top.connection_type_expressions) == 0
atomtype = AtomType()
atom1 = Atom(name='atom1', atom_type=atomtype)
top.add_site(atom1)
atom2 = Atom(name='atom2', atom_type=atomtype)
top.add_site(atom2)
assert top.n_sites == 2
assert len(top.atom_types) == 1
assert len(top.atom_type_expressions) == 1
assert top.n_connections == 0
assert len(top.connection_types) == 0
assert len(top.connection_type_expressions) == 0
ctype = BondType()
connection_12 = Bond(connection_members=[atom1, atom2],
bond_type=ctype)
top.add_connection(connection_12)
assert top.n_sites == 2
assert len(top.atom_types) == 1
assert len(top.atom_type_expressions) == 1
assert top.n_connections == 1
assert len(top.connection_types) == 1
assert len(top.connection_type_expressions) == 1
atom1.atom_type = AtomType(expression='sigma*epsilon*r')
assert top.n_sites == 2
assert len(top.atom_types) == 1
assert len(top.atom_type_expressions) == 1
assert top.n_connections == 1
assert len(top.connection_types) == 1
assert len(top.connection_type_expressions) == 1
top.update_atom_types()
assert top.n_sites == 2
assert len(top.atom_types) == 2
assert len(top.atom_type_expressions) == 2
assert top.n_connections == 1
assert len(top.connection_types) == 1
assert len(top.connection_type_expressions) == 1
def read_xyz(filename):
"""Reader for xyz file format.
Read in an xyz file at the given path and return a Topology object.
Parameters
----------
filename : str
Path to .xyz file that need to be read.
Return
------
top : topology.Topology
Topology object
"""
top = Topology()
with open(filename, "r") as xyz_file:
n_atoms = int(xyz_file.readline())
xyz_file.readline()
coords = np.zeros(shape=(n_atoms, 3)) * u.nanometer
for row, _ in enumerate(coords):
line = xyz_file.readline().split()
if not line:
msg = (
"Incorrect number of lines in input file. Based on the "
"number in the first line of the file, {} rows of atoms "
"were expected, but at least one fewer was found."
)
raise ValueError(msg.format(n_atoms))
tmp = np.array(line[1:4], dtype=np.float) * u.angstrom
coords[row] = tmp.in_units(u.nanometer)
site = Atom(name=line[0], position=coords[row])
top.add_site(site)
top.update_topology()
# Verify we have read the last line by ensuring the next line in blank
line = xyz_file.readline().split()
if line:
msg = (
"Incorrect number of lines in input file. Based on the "
"number in the first line of the file, {} rows of atoms "
"were expected, but at least one more was found."
)
raise ValueError(msg.format(n_atoms))
return top
def methane(self):
mytop = Topology()
c = Atom(name='c')
h1 = Atom(name='h1')
h2 = Atom(name='h2')
h3 = Atom(name='h3')
h4 = Atom(name='h4')
ch1 = Bond(connection_members=[c, h1])
ch2 = Bond(connection_members=[c, h2])
ch3 = Bond(connection_members=[c, h3])
ch4 = Bond(connection_members=[c, h4])
mytop.add_site(c, update_types=False)
mytop.add_site(h1, update_types=False)
mytop.add_site(h2, update_types=False)
mytop.add_site(h3, update_types=False)
mytop.add_site(h4, update_types=False)
mytop.add_connection(ch1, update_types=False)
mytop.add_connection(ch2, update_types=False)
mytop.add_connection(ch3, update_types=False)
mytop.add_connection(ch4, update_types=False)
mytop.update_topology()
return mytop
def from_mbuild(compound, box=None, search_method=element_by_symbol):
"""Convert an mbuild.Compound to a gmso.Topology
This conversion makes the following assumptions about the inputted
`Compound`:
* All positional and box dimension values in compound are in nanometers
* If the `Compound` has 4 or more levels of hierarchy, these are\
compressed to 3 levels of hierarchy in the resulting `Topology`. The\
top level `Compound` becomes the `Topology`, the second level\
Compounds become `SubTopologies`, and each particle becomes a `Site`,\
which are added to their corresponding `SubTopologies`.\
* Furthermore, `Sites` that do not belong to a sub-`Compound` are\
added to a single-`Site` `SubTopology`.
* The box dimension are extracted from `compound.periodicity`. If the\
`compound.periodicity` is `None`, the box lengths are the lengths of\
the bounding box + a 0.5 nm buffer.
* Only `Bonds` are added for each bond in the `Compound`. If `Angles`\
and `Dihedrals` are desired in the resulting `Topology`, they must be\
added separately from this function.
Parameters
----------
compound : mbuild.Compound
mbuild.Compound instance that need to be converted
box : mbuild.Box, optional, default=None
Box information to be loaded to a gmso.Topology
search_method : function, optional, default=element_by_symbol
Searching method used to assign element from periodic table to
particle site.
The information specified in the `search_method` argument is extracted
from each `Particle`'s `name` attribute.
Valid functions are element_by_symbol, element_by_name,
element_by_atomic_number, and element_by_mass, which can be imported
from `gmso.core.element'
Returns
-------
top : gmso.Topology
"""
msg = ("Argument compound is not an mbuild.Compound")
assert isinstance(compound, mb.Compound), msg
top = Topology()
top.typed = False
# Keep the name if it is not the default mBuild Compound name
if compound.name != mb.Compound().name:
top.name = compound.name
site_map = dict()
for child in compound.children:
if len(child.children) == 0:
continue
else:
subtop = SubTopology(name=child.name)
top.add_subtopology(subtop)
for particle in child.particles():
pos = particle.xyz[0] * u.nanometer
ele = search_method(particle.name)
site = Site(name=particle.name, position=pos, element=ele)
site_map[particle] = site
subtop.add_site(site)
top.update_topology()
for particle in compound.particles():
already_added_site = site_map.get(particle, None)
if already_added_site:
continue
pos = particle.xyz[0] * u.nanometer
ele = search_method(particle.name)
site = Site(name=particle.name, position=pos, element=ele)
site_map[particle] = site
# If the top has subtopologies, then place this particle into
# a single-site subtopology -- ensures that all sites are in the
# same level of hierarchy.
if len(top.subtops) > 0:
subtop = SubTopology(name=particle.name)
top.add_subtopology(subtop)
subtop.add_site(site)
else:
top.add_site(site)
for b1, b2 in compound.bonds():
new_bond = Bond(connection_members=[site_map[b1], site_map[b2]],
connection_type=None)
top.add_connection(new_bond)
top.update_topology()
if box:
top.box = from_mbuild_box(box)
# Assumes 2-D systems are not supported in mBuild
# if compound.periodicity is None and not box:
else:
if np.allclose(compound.periodicity, np.zeros(3)):
box = from_mbuild_box(compound.boundingbox)
if box:
box.lengths += [0.5, 0.5, 0.5] * u.nm
top.box = box
else:
top.box = Box(lengths=compound.periodicity)
return top
def read_gro(filename):
"""Provided a filepath to a gro file, generate a topology.
The Gromos87 (gro) format is a common plain text structure file used
commonly with the GROMACS simulation engine. This file contains the
simulation box parameters, number of atoms, the residue and atom number for
each atom, as well as their positions and velocities (velocity is
optional). This method will receive a filepath representation either as a
string, or a file object and return a `topology`.
Parameters
----------
filename : str or file object
The path to the gro file either as a string, or a file object that
points to the gro file.
Returns
-------
gmso.core.topology
A `topology` object containing site information
Notes
-----
Gro files do not specify connections between atoms, the returned topology
will not have connections between sites either.
Currently this implementation does not support a gro file with more than 1
frame.
All residues and resid information from the gro file are currently lost
when converting to `topology`.
"""
top = Topology()
with open(filename, 'r') as gro_file:
top.name = str(gro_file.readline().strip())
n_atoms = int(gro_file.readline())
coords = u.nm * np.zeros(shape=(n_atoms, 3))
for row, _ in enumerate(coords):
line = gro_file.readline()
if not line:
msg = (
'Incorrect number of lines in .gro file. Based on the '
'number in the second line of the file, {} rows of'
'atoms were expected, but at least one fewer was found.')
raise ValueError(msg.format(n_atoms))
resid = int(line[:5])
res_name = line[5:10]
atom_name = line[10:15]
atom_id = int(line[15:20])
coords[row] = u.nm * np.array([
float(line[20:28]),
float(line[28:36]),
float(line[36:44]),
])
site = Atom(name=atom_name, position=coords[row])
top.add_site(site, update_types=False)
top.update_topology()
# Box information
line = gro_file.readline().split()
top.box = Box(u.nm * np.array([float(val) for val in line[:3]]))
# Verify we have read the last line by ensuring the next line in blank
line = gro_file.readline()
if line:
msg = ('Incorrect number of lines in input file. Based on the '
'number in the second line of the file, {} rows of atoms '
'were expected, but at least one more was found.')
raise ValueError(msg.format(n_atoms))
return top
def read_lammpsdata(filename, atom_style='full'):
top = Topology()
# The idea is to get the number of types to read in
with open(filename, 'r') as lammps_file:
for i, line in enumerate(lammps_file):
if 'xlo' in line.split():
break
typelines = open(filename, 'r').readlines()[2:i]
# TODO: Rewrite the logic to read in all type information
for line in typelines:
if 'atoms' in line:
n_atoms = int(line.split()[0])
elif 'bonds' in line:
n_bonds = int(line.split()[0])
elif 'angles' in line:
n_angles = int(line.split()[0])
elif 'dihedrals' in line:
n_dihedrals = int(line.split()[0])
elif 'impropers' in line:
n_impropers = int(line.split()[0])
elif 'atom' in line:
n_atomtypes = int(line.split()[0])
elif 'bond' in line:
n_bondtypes = int(line.split()[0])
elif 'angle' in line:
n_angletypes = int(line.split()[0])
elif 'dihedral' in line:
n_dihedraltypes = int(line.split()[0])
with open(filename, 'r') as lammps_file:
top.name = str(lammps_file.readline().strip())
lammps_file.readline()
for j in range(
i -
2): # looping through to skip through all of the type lines
lammps_file.readline()
x_line = lammps_file.readline().split()
y_line = lammps_file.readline().split()
z_line = lammps_file.readline().split()
x = float(x_line[1]) - float(x_line[0])
y = float(y_line[1]) - float(y_line[0])
z = float(z_line[1]) - float(z_line[0])
# Box Information
lengths = u.unyt_array([x, y, z], u.angstrom)
top.box = Box(lengths)
coords = u.angstrom * np.zeros(shape=(n_atoms, 3))
unique_types = _get_masses(filename, n_atomtypes)
charge_dict, coords_dict, type_dict = _get_atoms(filename, n_atoms, coords)
sigma_dict, epsilon_dict = _get_pairs(filename, n_atomtypes)
for k, v in type_dict.items():
atomtype = AtomType(name=k,
mass=unique_types[k],
charge=charge_dict[k],
parameters={
'sigma': sigma_dict[k],
'epsilon': epsilon_dict[k]
})
for i in range(v):
site = Site(
name="atom{}".format(i), # probably change
position=coords_dict[k][i],
atom_type=atomtype)
top.add_site(site, update_types=False)
print('{}:{}'.format(k, i))
top.update_topology()
return top
def read_lammpsdata(filename,
atom_style="full",
unit_style="real",
potential="lj"):
"""Read in a lammps data file as a GMSO topology.
Parameters
----------
filename : str
LAMMPS data file
atom_style : str, optional, default='full'
Inferred atom style defined by LAMMPS
potential: str, optional, default='lj'
Potential type defined in data file
Returns
-------
top : GMSO Topology
A GMSO Topology object
Notes
-----
See http://lammps.sandia.gov/doc/2001/data_format.html for a full description of the LAMMPS data format.
This is a work in progress, as only several atom styles, potential styles, and unit styes are currently supported.
Currently supporting the following atom styles: 'full'
Currently supporting the following unit styles: 'real'
Currently supporting the following potential styles: 'lj'
Proper dihedrals can be read in but is currently not tested.
Currently not supporting improper dihedrals.
"""
# TODO: Add argument to ask if user wants to infer bond type
top = Topology()
# Validate 'atom_style'
if atom_style not in ["full"]:
raise ValueError(
'Atom Style "{}" is invalid or is not currently supported'.format(
atom_style))
# Validate 'unit_style'
if unit_style not in ["real"]:
raiseValueError(
'Unit Style "{}" is invalid or is not currently supported'.format(
unit_style))
# Parse box information
_get_box_coordinates(filename, unit_style, top)
# Parse atom type information
top, type_list = _get_ff_information(filename, unit_style, top)
# Parse atom information
_get_atoms(filename, top, unit_style, type_list)
# Parse connection (bonds, angles, dihedrals) information
# TODO: Add more atom styles
if atom_style in ["full"]:
_get_connection(filename, top, unit_style, connection_type="bond")
_get_connection(filename, top, unit_style, connection_type="angle")
top.update_topology()
return top
