def write_molecule(
path,
component: Component,
remove_hs: bool = True,
alt_names: bool = False,
conf_type: ConformerType = ConformerType.Ideal,
):
"""Export molecule in a specified format. Presently supported formats
are: PDB CCD CIF (*.cif); Mol file (*.sdf); Chemical Markup language
(*.cml); PDB file (*.pdb); XYZ file (*.xyz); XML (*.xml).
ConformerType.AllConformers is presently supported only for PDB.
Args:
path (str|Path): Path to the file. Suffix determines format to be
used.
component (Component): Component to be exported
remove_hs (bool, optional): Defaults to True. Whether or not
hydrogens should be removed.
alt_names (bool, optional): Defaults to False. Whether or not
alternate names should be exported.
conf_type (ConformerType, optional):
Defaults to ConformerType.Ideal. Conformer type to be
exported.
Raises:
CCDUtilsError: For unsupported format
"""
path = str(path) if isinstance(path, Path) else path
extension = path.split(".")[-1].lower()
str_representation = ""
if extension in ("sdf", "mol"):
str_representation = to_sdf_str(component, remove_hs, conf_type)
elif extension == "pdb":
str_representation = to_pdb_str(component, remove_hs, alt_names,
conf_type)
elif extension in ("mmcif", "cif"):
to_pdb_ccd_cif_file(path, component, remove_hs)
return
elif extension == "cml":
str_representation = to_cml_str(component, remove_hs, conf_type)
elif extension == "xml":
str_representation = to_xml_str(component, remove_hs, conf_type)
elif extension == "xyz":
str_representation = to_xyz_str(component, remove_hs, conf_type)
elif extension == "json":
str_representation = json.dumps(to_json_dict(component, remove_hs,
conf_type),
sort_keys=True,
indent=4)
else:
raise CCDUtilsError("Unsupported file format: {}".format(extension))
with open(path, "w") as f:
f.write(str_representation)
def to_sdf_str(
component: Component,
remove_hs: bool = True,
conf_type: ConformerType = ConformerType.Ideal,
):
"""Converts structure to the SDF format.
Args:
component (Component): Component to be exported.
remove_hs (bool, optional): Defaults to True.
conf_type (ConformerType, optional): Defaults to ConformerType.Ideal.
Raises:
CCDUtilsError: In case the structure could not be exported.
Returns:
str: String representation of the component in the SDF format
"""
(mol_to_save, _, conf_type) = _prepate_structure(component, remove_hs,
conf_type)
mol_block = []
if conf_type == ConformerType.AllConformers:
conformers = [
ConformerType.Model, ConformerType.Ideal, ConformerType.Computed
]
else:
conformers = [conf_type]
try:
for c in conformers:
try:
conf_id = -1
if c != ConformerType.AllConformers:
conf_id = component.get_conformer(c).GetId()
block = [
f"{component.id} - {c.name} conformer",
rdkit.Chem.MolToMolBlock(mol_to_save,
confId=conf_id).strip(),
"$$$$\n",
]
mol_block += block
except ValueError as e:
if str(e) == "Bad Conformer Id":
pass
else:
raise CCDUtilsError(f"Error writing SDF file - {e}")
except Exception:
mol_block = _to_sdf_str_fallback(mol_to_save, component.id, conformers)
return "\n".join(mol_block)
def write_molecule(path,
component: Component,
remove_hs: bool = True,
alt_names: bool = False,
conf_type: ConformerType = ConformerType.Ideal):
"""Export molecule in a specified format. Presently supported formats
are: PDB CCD CIF (*.cif); Mol file (*.sdf); Chemical Markup language
(*.cml); PDB file (*.pdb); XYZ file (*.xyz); XML (*.xml).
ConformerType.AllConformers is presently supported only for PDB.
Args:
path (str): Path to the file. Extension determines format to be
used.
component (Component): Component to be exported
remove_hs (bool, optional): Defaults to True. Whether or not
hydrogens should be removed.
alt_names (bool, optional): Defaults to False. Whether or not
alternate names should be exported.
conf_type (ConformerType, optional):
Defaults to ConformerType.Ideal. Conformer type to be
exported.
Raises:
CCDUtilsError: For unsupported format
"""
extension = path.split('.')[-1].lower()
str_representation = ''
if extension in ('sdf', 'mol'):
str_representation = to_sdf_str(component, remove_hs, conf_type)
elif extension == 'pdb':
str_representation = to_pdb_str(component, remove_hs, alt_names,
conf_type)
elif extension in ('mmcif', 'cif'):
to_pdb_ccd_cif_file(path, component, remove_hs)
return
elif extension == 'cml':
str_representation = to_cml_str(component, remove_hs, conf_type)
elif extension == 'xml':
str_representation = to_xml_str(component, remove_hs, conf_type)
elif extension == 'xyz':
str_representation = to_xyz_str(component, remove_hs, conf_type)
elif extension == 'json':
str_representation = json.dumps(to_json_dict(component, remove_hs,
conf_type),
sort_keys=True,
indent=4)
else:
raise CCDUtilsError('Unsupported file format: {}'.format(extension))
with open(path, 'w') as f:
f.write(str_representation)
def to_sdf_str(component: Component,
remove_hs: bool = True,
conf_type: ConformerType = ConformerType.Ideal):
"""Converts structure to the SDF format.
Args:
component (Component): Component to be exported.
remove_hs (bool, optional): Defaults to True.
conf_type (ConformerType, optional): Defaults to ConformerType.Ideal.
Raises:
CCDUtilsError: In case the structure could not be exported.
Returns:
str: String representation of the component in the SDF format
"""
(mol_to_save, conf_id,
conf_type) = _prepate_structure(component, remove_hs, conf_type)
mol_block = []
mappings = {}
if conf_type == ConformerType.AllConformers:
conformers = [
ConformerType.Model, ConformerType.Ideal, ConformerType.Computed
]
else:
conformers = [conf_type]
try:
for conf in conformers:
try:
s = '{} - {} conformer'.format(component.id, conf.name)
s += rdkit.Chem.MolToMolBlock(
mol_to_save, confId=component.conformers_mapping[conf])
s += '$$$$'
mol_block.append(s)
except ValueError as e:
if str(e) == 'Bad Conformer Id':
pass
else:
raise CCDUtilsError(
'Error writing SDF file - {}'.format(e))
except Exception:
mappings = {
m.name: component.conformers_mapping[m]
for m in conformers
}
mol_block = _to_sdf_str_fallback(mol_to_save, component.id, mappings)
return "\n".join(mol_block)
def get_scaffolds(self,
scaffolding_method=ScaffoldingMethod.MurckoScaffold):
"""Compute deemed scaffolds for a given compound.
Args:
scaffolding_method (ScaffoldingMethod, optional):
Defaults to MurckoScaffold. Scaffolding method to use
Returns:
list[rdkit.Chem.rdchem.Mol]: Scaffolds found in the component.
"""
try:
scaffolds = []
if scaffolding_method == ScaffoldingMethod.MurckoScaffold:
scaffolds = [(MurckoScaffold.GetScaffoldForMol(self.mol_no_h))]
elif scaffolding_method == ScaffoldingMethod.MurckoGeneric:
scaffolds = [
(MurckoScaffold.MakeScaffoldGeneric(self.mol_no_h))
]
elif scaffolding_method == ScaffoldingMethod.Brics:
scaffolds = BRICS.BRICSDecompose(self.mol_no_h)
brics_smiles = [
re.sub(r"(\[[0-9]*\*\])", "[H]", i) for i in scaffolds
] # replace dummy atoms with H's to get matches https://sourceforge.net/p/rdkit/mailman/message/35261974/
brics_mols = [
rdkit.Chem.MolFromSmiles(x) for x in brics_smiles
]
for mol in brics_mols:
rdkit.Chem.RemoveHs(mol)
brics_hits = [
self.mol_no_h.GetSubstructMatches(i) for i in brics_mols
]
for index, brics_hit in enumerate(brics_hits):
smiles = rdkit.Chem.MolToSmiles(brics_mols[index])
name = scaffolding_method.name
source = 'RDKit scaffolds'
key = f'{name}_{smiles}'
brics_hit = conversions.listit(brics_hit)
if not smiles:
continue
if key not in self._scaffolds:
self._scaffolds[key] = SubstructureMapping(
name, smiles, source, brics_hit)
return brics_mols
for s in scaffolds:
scaffold_atom_names = [
atom.GetProp('name') for atom in s.GetAtoms()
]
mapping = []
for at_name in scaffold_atom_names:
idx = [
atom.GetIdx() for atom in self.mol.GetAtoms()
if atom.GetProp('name') == at_name
][0]
mapping.append(idx)
smiles = rdkit.Chem.MolToSmiles(s)
name = scaffolding_method.name
source = 'RDKit scaffolds'
if not smiles:
continue
if name in self._scaffolds:
self._scaffolds[name].mappings.append(mapping)
else:
self._scaffolds[name] = SubstructureMapping(
name, smiles, source, [mapping])
return scaffolds
except (RuntimeError, ValueError):
raise CCDUtilsError(
f'Computing scaffolds using method {scaffolding_method.name} failed.'
)
def export_2d_svg(self,
file_name: str,
width: int = 500,
names: bool = False,
wedge_bonds: bool = True,
atom_highlight: Dict[Any, Tuple] = None,
bond_highlight: Dict[Tuple, Tuple] = None):
"""Save 2D depiction of the component as an SVG file. Component
id is generated in case the image cannot be drawn.
Args:
file_name (str): path to store 2d depiction
width (int, optional): Defaults to 500. Width of a frame in pixels.
names (bool, optional): Defaults to False. Whether or not to
include atom names in depiction. If atom name is not set, element symbol is used instead.
wedge_bonds (bool, optional): Defaults to True. Whether or not
the molecule should be depicted with bond wedging.
atomHighlight (:obj:`dict` of :obj:`tuple` of :obj:`float`, optional):
Defaults to None. Atoms names to be highlighted along
with colors in RGB. e.g. {'CA': (0.5, 0.5, 0.5)} or {0: (0.5, 0.5, 0.5)}
bondHighlight (:obj:`dict` of :obj:`tuple` of :obj:`float`, optional):
Defaults to None. Bonds to be highlighted along with
colors in RGB. e.g. {('CA', 'CB'): (0.5, 0.5, 0.5)} or {(0, 1): (0.5, 0.5, 0.5)}
Raises:
CCDUtilsError: If bond or atom does not exist.
"""
if self.mol2D is None:
drawing.save_no_image(file_name, self.id, width)
return
drawer = Draw.rdMolDraw2D.MolDraw2DSVG(width, width)
options = drawer.drawOptions()
atom_mapping = {
self._get_atom_name(a): i
for i, a in enumerate(self.mol2D.GetAtoms())
}
atom_highlight = {} if atom_highlight is None else atom_highlight
bond_highlight = {} if bond_highlight is None else bond_highlight
if width < 201:
options.bondLineWidth = 1
if all(isinstance(i, str) for i in atom_highlight.keys()):
atom_highlight = {
atom_mapping[k]: v
for k, v in atom_highlight.items()
}
else:
atom_highlight = {}
if bond_highlight:
if all(
isinstance(i[0], str) and isinstance(i[1], str)
for i in bond_highlight.keys()):
temp_highlight = {}
for k, v in bond_highlight.items():
bond = self.mol2D.GetBondBetweenAtoms(
atom_mapping[k[0]], atom_mapping[k[1]])
if bond is None:
raise CCDUtilsError(
'Bond between {} and {} does not exist'.format(
k[0], k[1]))
temp_highlight[bond.GetIdx()] = v
bond_highlight = temp_highlight
if names:
for i, a in enumerate(self.mol2D.GetAtoms()):
atom_name = self._get_atom_name(a)
options.atomLabels[i] = atom_name
a.SetProp('molFileAlias', atom_name)
drawing.draw_molecule(self.mol2D, drawer, file_name, wedge_bonds,
atom_highlight, bond_highlight)