def test_from_smiles(smiles):
"""
Make sure hydrogens are added to a molecule when needed.
"""
mol = Ligand.from_smiles(smiles_string=smiles, name="methane")
# count the number of hydrogens
hs = sum([1 for atom in mol.atoms if atom.atomic_symbol == "H"])
assert hs == 4
def test_torsion_finder_multiple():
"""
Find non hydrogen torsions for multiple rotatable bonds.
"""
mol = Ligand.from_smiles("CCO", "ethanol")
bonds = mol.find_rotatable_bonds()
for bond in bonds:
torsion = find_heavy_torsion(molecule=mol, bond=bond)
check_proper_torsion(torsion=torsion, molecule=mol)
def test_to_rdkit_complicated_stereo():
"""
Make sure we can convert a complicated molecule with multiple stereo centres to rdkit.
"""
mol = Ligand.from_smiles(
"[H][C@]1([C@@]([C@](O[C@@]1([H])C([H])([H])OP(=O)(O[H])O[H])([H])N2C(=C([N+](C2([H])[H])([H])[H])C(=O)N([H])[H])O[H])([H])O[H])O[H]",
name="complicated",
)
mol.to_rdkit()
def test_torsion_special_case_double():
"""
Make sure special cases changes the scan range for a bond.
"""
mol = Ligand.from_smiles("CO", "methanol")
t_scan = TorsionScan1D()
t_scan.clear_avoided_torsions()
# add the special case with non-default range
t_scan.add_special_torsion(smirks="[*:1]-[OH1:2]", scan_range=(0, 180))
# get the one rotatable bond
bond = mol.find_rotatable_bonds()[0]
scan_range = t_scan._get_scan_range(molecule=mol, bond=bond)
assert scan_range == (0, 180)
def test_to_rdkit_kekule(molecule):
"""
Make sure we can correctly convert a molecule to rdkit when it has mixed aromatic and non-aromatic rings.
This test is due to an kekule error where non aromatic atoms were sometimes given aromatic bonds.
"""
mol = Ligand.from_smiles(molecule, name="test")
rd_mol = mol.to_rdkit()
for atom in rd_mol.GetAtoms():
qb_atom = mol.atoms[atom.GetIdx()]
assert atom.GetIsAromatic() is qb_atom.aromatic
for bond in rd_mol.GetBonds():
qb_bond = mol.bonds[bond.GetIdx()]
assert qb_bond.aromatic is bond.GetIsAromatic()
assert qb_bond.bond_order == bond.GetBondTypeAsDouble()
def run(
input_file: Optional[str] = None,
smiles: Optional[str] = None,
name: Optional[str] = None,
multiplicity: int = 1,
end: Optional[str] = None,
skip_stages: Optional[List[str]] = None,
config: Optional[str] = None,
protocol: Optional[str] = None,
cores: Optional[int] = None,
memory: Optional[int] = None,
):
"""Run the QUBEKit parametrisation workflow on an input molecule."""
# make sure we have an input or smiles not both
if input_file is not None and smiles is not None:
raise RuntimeError(
"Please supply either the name of the input file or a smiles string not both."
)
# load the molecule
if input_file is not None:
molecule = Ligand.from_file(file_name=input_file,
multiplicity=multiplicity)
else:
if name is None:
raise RuntimeError(
"Please also pass a name for the molecule when starting from smiles."
)
molecule = Ligand.from_smiles(smiles_string=smiles,
name=name,
multiplicity=multiplicity)
# load workflow
workflow = prep_config(config_file=config,
memory=memory,
cores=cores,
protocol=protocol)
# move into the working folder and run
with folder_setup(
f"QUBEKit_{molecule.name}_{datetime.now().strftime('%Y_%m_%d')}"):
# write the starting molecule
molecule.to_file(file_name=f"{molecule.name}.pdb")
workflow.new_workflow(molecule=molecule,
skip_stages=skip_stages,
end=end)
def test_double_dihedral(tmpdir):
"""Test running a molecule with two rotatable bonds."""
with tmpdir.as_cwd():
mol = Ligand.from_smiles("CCO", "ethanol")
# build a scanner with grid spacing 60 and clear out avoided methyl
tdrive = TorsionDriver(
program="rdkit",
method="uff",
basis=None,
cores=1,
memory=1,
n_workers=1,
grid_spacing=60,
)
t_scan = TorsionScan1D(torsion_driver=tdrive)
t_scan.clear_avoided_torsions()
result_mol = t_scan.run(molecule=mol)
assert len(result_mol.qm_scans) == 2
# make sure input molecule coords were not changed
assert np.allclose(mol.coordinates, result_mol.coordinates)
def mol_47():
return Ligand.from_smiles("CC(C)(O)CCC(C)(C)O", "mol_47")
def run(
bulk_file: str,
skip_stages: Optional[List[str]] = None,
end: Optional[str] = None,
restart: Optional[str] = None,
config: Optional[str] = None,
protocol: Optional[str] = None,
cores: Optional[int] = None,
memory: Optional[int] = None,
) -> None:
"""Run the QUBEKit parametrisation workflow on a collection of molecules in serial.
Loop over the molecules in order of the CSV file.
"""
import glob
import os
from qubekit.utils.helpers import mol_data_from_csv
home = os.getcwd()
# load all inputs
bulk_data = mol_data_from_csv(bulk_file)
# start main molecule loop
for name, mol_data in bulk_data.items():
print(f"Analysing: {name}")
try:
if restart is not None or mol_data["restart"] is not None:
# we are trying to restart a run, find the folder
# should only be one
fname = name.split(".")[0]
folder = glob.glob(f"QUBEKit_{fname}_*")[0]
with folder_setup(folder):
results = WorkFlowResult.parse_file("workflow_result.json")
if config is None:
# if we have no new config load from results
workflow = prep_config(
results=results, cores=cores, memory=memory, protocol=None
)
else:
# load the new config file
workflow = prep_config(
config_file=config, cores=cores, memory=memory
)
workflow.restart_workflow(
start=restart or mol_data["restart"],
skip_stages=skip_stages,
end=end or mol_data["end"],
result=results,
)
else:
if mol_data["smiles"] is not None:
molecule = Ligand.from_smiles(
smiles_string=mol_data["smiles"], name=name
)
else:
molecule = Ligand.from_file(file_name=name)
# load the CLI config or the csv config, else default
workflow = prep_config(
config_file=config or mol_data["config_file"],
memory=memory,
cores=cores,
protocol=protocol,
)
# move into the working folder and run
with folder_setup(
f"QUBEKit_{molecule.name}_{datetime.now().strftime('%Y_%m_%d')}"
):
# write the starting molecule
molecule.to_file(file_name=f"{molecule.name}.pdb")
workflow.new_workflow(
molecule=molecule,
skip_stages=skip_stages,
end=end or mol_data["end"],
)
except WorkFlowExecutionError:
os.chdir(home)
print(
f"An error was encountered while running {name} see folder for more info."
)
continue
