def __init__(self, topologies):
"""
Initializes an OBC1 object.
Parameters
----------
topologies : a Topology object or list[Topology object]
The molecular topology representation to write as a
Impact template
"""
# Not implemented in PELE
logger = Logger()
logger.warning('OBC1 is not implemented in PELE')
super().__init__(topologies)
def _build(self):
"""The topology builder."""
# In case the molecule has not been initialized
if (self.molecule.rdkit_molecule is None
or len(list(self.parameters.atom_iterator)) == 0):
logger = Logger()
logger.warning('Warning: the input molecule has not been ' +
' initialized and its topology will be empty')
return
self._build_atoms()
self._build_bonds()
self._build_angles()
self._build_propers()
self._build_impropers()
def _pdb_checkup(self, path):
"""
Safety check for PDB files in order to properly handle exceptions
related with its format prior running the parser.
Parameters
----------
path : str
The path to a PDB with the molecule structure
"""
# Parse PDB file
atom_id, res_name, res_id = ([] for i in range(3))
connectivity = False
with open(path) as pdb_file:
for line in pdb_file:
if line.startswith('ATOM') or line.startswith('HETATM'):
atom_id.append(line[12:16])
res_name.append(line[17:20])
res_id.append(line[22:26])
if line.startswith('CONECT'):
connectivity = True
# Handle exceptions related with the PDB file format
if not res_id[:-1] == res_id[1:]:
raise Exception(
'A single ligand with immutable residue ids is expected')
if not res_name[:-1] == res_name[1:]:
raise Exception(
'A single ligand with immutable residue names is expected')
if not len(atom_id) == len(set(atom_id)):
raise Exception('Ligand in input PDB has no unique atom names')
if not connectivity and self.connectivity_template is None:
log = Logger()
log.warning(
"Warning: input PDB has no information about the " +
"connectivity and this could result in an unexpected " +
"bond assignment")
def _extract_molecules_from_chain(self, chain, rotamer_resolution,
exclude_terminal_rotamers,
allow_undefined_stereo,
core_constraints):
"""
It extracts all hetero molecules found in the selected the chain
of a PDB file.
Parameters
----------
chain_id : str
Chain ID.
rotamer_resolution : float
The resolution in degrees to discretize the rotamer's
conformational space. Default is 30
exclude_terminal_rotamers : bool
Whether to exclude terminal rotamers when generating the
rotamers library or not
allow_undefined_stereo : bool
Whether to allow a molecule with undefined stereochemistry
to be defined or try to assign the stereochemistry and
raise a complaint if not possible. Default is False
core_constraints : list[int or str]
It defines the list of atoms to constrain in the core, thus,
the core will be forced to contain them. Atoms can be specified
through integers that match the atom index or strings that
match with the atom PDB name
Returns
-------
molecules : list[peleffy.topology.Molecule object]
Selected molecules
"""
from peleffy.topology.molecule import Molecule
# Check if there is more than one hetero molecule in the same chain
residues_ids = set([
line[22:26].strip() for line in self.pdb_content
if line.startswith('HETATM') and line[21:22] == chain
and not line[17:20].strip() == 'HOH'
])
molecules = []
for residue_id in residues_ids:
res_name = set([
line[17:20].strip() for line in self.pdb_content
if line.startswith('HETATM') and line[21:22] == chain
and line[22:26].strip() == residue_id
])
# Select which atoms compose this hetero molecule
atom_ids = [
line[6:11].strip() for line in self.pdb_content
if line.startswith('HETATM') and line[21:22] == chain
and line[22:26].strip() == residue_id
]
# Extract the PDB block of the molecule
pdb_block = [
line for line in self.pdb_content
if (line.startswith('HETATM') or line.startswith('CONECT'))
and any(' {} '.format(a) in line for a in atom_ids)
]
try:
molecules.append(
Molecule(
pdb_block=''.join(pdb_block),
rotamer_resolution=rotamer_resolution,
exclude_terminal_rotamers=exclude_terminal_rotamers,
allow_undefined_stereo=allow_undefined_stereo,
core_constraints=core_constraints))
except Exception as e:
log = Logger()
log.warning(' - Skipping {} '.format(list(res_name)[0]) +
'from chain {}'.format(chain))
log.warning(' - The following exception was raised: ' +
'{}'.format(e))
return molecules
def _read_and_fix_pdb(self, path):
"""
It reads the input PDB file returns the corresponding PDB block.
It also applies some modifications, in case it requires some
fixing prior running the parser.
Parameters
----------
path : str
The path to a PDB with the molecule structure
Returns
-------
pdb_block : str
The corresponding PDB block, with applied fixes if required
"""
log = Logger()
# Skip PDB fixing if it has been deactivated
if not self.fix_pdb:
with open(path) as pdb_file:
pdb_block = pdb_file.read()
return pdb_block
# Fix PDB
missing_element = False
any_fail = False
pdb_block = ''
with open(path) as pdb_file:
for line in pdb_file:
if line.startswith('ATOM') or line.startswith('HETATM'):
if len(line) < 78 or line[76:78] == ' ':
missing_element = True
atom_name = line[12:16]
# Try to infer element from atom name
inferred_element = ''.join([
c for c in atom_name
if not c.isdigit() and c != ' '
])
# Format properly the element identifier
if len(inferred_element) == 1:
inferred_element = inferred_element.upper()
elif len(inferred_element) == 2:
inferred_element = inferred_element[0].upper() + \
inferred_element[1].lower()
else:
# We were expecting an element identifier of 1 or 2 chars
any_fail = True
break
# Remove line breaks, if any
line = line.strip()
# Fill a short line with white spaces
while (len(line) < 79):
line += ' '
# Add element to line (right-justified)
line = line[:76] + '{:>2s}'.format(inferred_element) \
+ line[79:] + '\n'
pdb_block += line
if missing_element:
log.warning("Warning: input PDB has no information about atom " +
"elements and they were inferred from atom names. " +
"Please, verify that the resulting elements are " +
"correct")
if any_fail:
log.error("Error: PDB could not be fixed")
with open(path) as pdb_file:
pdb_block = pdb_file.read()
return pdb_block