def topology(self):
"""Get the topology out from the file
Returns
-------
topology : mdtraj.Topology
A topology object
"""
try:
raw = self._get_node('/', name='topology')[0]
if not isinstance(raw, string_types):
raw = raw.decode()
topology_dict = json.loads(raw)
except self.tables.NoSuchNodeError:
return None
topology = Topology()
for chain_dict in sorted(topology_dict['chains'],
key=operator.itemgetter('index')):
chain = topology.add_chain()
for residue_dict in sorted(chain_dict['residues'],
key=operator.itemgetter('index')):
try:
resSeq = residue_dict["resSeq"]
except KeyError:
resSeq = None
warnings.warn(
'No resSeq information found in HDF file, defaulting to zero-based indices'
)
residue = topology.add_residue(residue_dict['name'],
chain,
resSeq=resSeq)
for atom_dict in sorted(residue_dict['atoms'],
key=operator.itemgetter('index')):
try:
element = elem.get_by_symbol(atom_dict['element'])
except KeyError:
element = None
topology.add_atom(atom_dict['name'], element, residue)
atoms = list(topology.atoms)
for index1, index2 in topology_dict['bonds']:
topology.add_bond(atoms[index1], atoms[index2])
return topology
def load_coor(filename):
"""
Load a simple coordinate file, formatted as:
x y z atomic_number
where `x`,`y`,`z` (float) are the positions of the atom, in angstroms, and
`atomic_number` (int) is the atomic number Z specifying what the atom is.
Parameters
----------
filename : str
The filename to load.
Returns
-------
structure : mdtraj.trajectory
A meta-data minimal mdtraj instance
"""
data = np.genfromtxt(filename)
xyz = data[:, :3] / 10.0 # coor files are in angstoms, conv. to nm
atomic_numbers = data[:, 3]
top = Topology()
chain = top.addChain()
residue = top.addResidue('XXX', chain)
for i in range(data.shape[0]):
element_symb = periodic_table[atomic_numbers[i]][
1] # should give symbol
element = Element.getBySymbol(element_symb)
name = '%s' % element_symb
top.addAtom(name, element, residue)
structure = trajectory.Trajectory(xyz=xyz, topology=top)
return structure
def _read_models(self):
if not self._mode == 'r':
raise ValueError('file not opened for reading')
self._topology = Topology()
pdb = PdbStructure(self._file, load_all_models=True)
atomByNumber = {}
for chain in pdb.iter_chains():
c = self._topology.add_chain()
for residue in chain.iter_residues():
resName = residue.get_name()
if resName in PDBTrajectoryFile._residueNameReplacements:
resName = PDBTrajectoryFile._residueNameReplacements[resName]
r = self._topology.add_residue(resName, c, residue.number)
if resName in PDBTrajectoryFile._atomNameReplacements:
atomReplacements = PDBTrajectoryFile._atomNameReplacements[resName]
else:
atomReplacements = {}
for atom in residue.atoms:
atomName = atom.get_name()
if atomName in atomReplacements:
atomName = atomReplacements[atomName]
atomName = atomName.strip()
element = atom.element
if element is None:
element = self._guess_element(atomName, residue)
newAtom = self._topology.add_atom(atomName, element, r)
atomByNumber[atom.serial_number] = newAtom
# load all of the positions (from every model)
_positions = []
for model in pdb.iter_models(use_all_models=True):
coords = []
for chain in model.iter_chains():
for residue in chain.iter_residues():
for atom in residue.atoms:
coords.append(atom.get_position())
_positions.append(coords)
self._positions = np.array(_positions)
## The atom positions read from the PDB file
self._unitcell_lengths = pdb.get_unit_cell_lengths()
self._unitcell_angles = pdb.get_unit_cell_angles()
self._topology.create_standard_bonds()
self._topology.create_disulfide_bonds(self.positions[0])
# Add bonds based on CONECT records.
connectBonds = []
for connect in pdb.models[0].connects:
i = connect[0]
for j in connect[1:]:
connectBonds.append((atomByNumber[i], atomByNumber[j]))
if len(connectBonds) > 0:
# Only add bonds that don't already exist.
existingBonds = set(self._topology.bonds)
for bond in connectBonds:
if bond not in existingBonds and (bond[1], bond[0]) not in existingBonds:
self._topology.add_bond(bond[0], bond[1])
existingBonds.add(bond)
