class TestAmberToolsToolkitWrapper:
"""Test the AmberToolsToolkitWrapper"""
@pytest.mark.skipif(
not RDKitToolkitWrapper.is_available()
or not AmberToolsToolkitWrapper.is_available(),
reason='RDKitToolkit and AmberToolsToolkit not available')
def test_compute_partial_charges(self):
"""Test OpenEyeToolkitWrapper compute_partial_charges()"""
toolkit_registry = ToolkitRegistry(
toolkit_precedence=[AmberToolsToolkitWrapper, RDKitToolkitWrapper])
smiles = '[H]C([H])([H])C([H])([H])[H]'
molecule = Molecule.from_smiles(smiles,
toolkit_registry=toolkit_registry)
molecule.generate_conformers(toolkit_registry=toolkit_registry)
# TODO: Implementation of these tests is pending a decision on the API for our charge model
with pytest.raises(NotImplementedError) as excinfo:
charge_model = 'notARealChargeModel'
molecule.compute_partial_charges(toolkit_registry=toolkit_registry
) #, charge_model=charge_model)
# ['cm1', 'cm2']
for charge_model in ['gas', 'mul', 'bcc']:
with pytest.raises(NotImplementedError) as excinfo:
molecule.compute_partial_charges(
toolkit_registry=toolkit_registry
) #, charge_model=charge_model)
charge_sum = 0 * unit.elementary_charge
for pc in molecule._partial_charges:
charge_sum += pc
assert charge_sum < 0.01 * unit.elementary_charge
# For now, just test AM1-BCC while the SMIRNOFF spec for other charge models gets worked out
molecule.compute_partial_charges_am1bcc(
toolkit_registry=toolkit_registry) # , charge_model=charge_model)
charge_sum = 0 * unit.elementary_charge
for pc in molecule._partial_charges:
charge_sum += pc
assert charge_sum < 0.002 * unit.elementary_charge
@pytest.mark.skipif(
not RDKitToolkitWrapper.is_available()
or not AmberToolsToolkitWrapper.is_available(),
reason='RDKitToolkit and AmberToolsToolkit not available')
def test_compute_partial_charges_net_charge(self):
"""Test OpenEyeToolkitWrapper compute_partial_charges() on a molecule with a net +1 charge"""
toolkit_registry = ToolkitRegistry(
toolkit_precedence=[AmberToolsToolkitWrapper, RDKitToolkitWrapper])
smiles = '[H]C([H])([H])[N+]([H])([H])[H]'
molecule = Molecule.from_smiles(smiles,
toolkit_registry=toolkit_registry)
molecule.generate_conformers(toolkit_registry=toolkit_registry)
with pytest.raises(NotImplementedError) as excinfo:
charge_model = 'notARealChargeModel'
molecule.compute_partial_charges(toolkit_registry=toolkit_registry
) #, charge_model=charge_model)
# TODO: Figure out why ['cm1', 'cm2'] fail
for charge_model in ['gas', 'mul', 'bcc']:
with pytest.raises(NotImplementedError) as excinfo:
molecule.compute_partial_charges(
toolkit_registry=toolkit_registry
) #, charge_model=charge_model)
charge_sum = 0 * unit.elementary_charge
for pc in molecule._partial_charges:
charge_sum += pc
assert 0.99 * unit.elementary_charge < charge_sum < 1.01 * unit.elementary_charge
# For now, I'm just testing AM1-BCC (will test more when the SMIRNOFF spec for other charges is finalized)
molecule.compute_partial_charges_am1bcc(
toolkit_registry=toolkit_registry)
charge_sum = 0 * unit.elementary_charge
for pc in molecule._partial_charges:
charge_sum += pc
assert 0.999 * unit.elementary_charge < charge_sum < 1.001 * unit.elementary_charge
def test_to_from_rdkit_core_props_filled(self):
"""Test RDKitToolkitWrapper to_rdkit() and from_rdkit() when given populated core property fields"""
toolkit_wrapper = RDKitToolkitWrapper()
# Replacing with a simple molecule with stereochemistry
input_smiles = r'C\C(F)=C(/F)C[C@@](C)(Cl)Br'
expected_output_smiles = r'[H][C]([H])([H])/[C]([F])=[C](\[F])[C]([H])([H])[C@@]([Cl])([Br])[C]([H])([H])[H]'
molecule = Molecule.from_smiles(input_smiles,
toolkit_registry=toolkit_wrapper)
assert molecule.to_smiles(
toolkit_registry=toolkit_wrapper) == expected_output_smiles
# Populate core molecule property fields
molecule.name = 'Alice'
partial_charges = unit.Quantity(
np.array([
-.9, -.8, -.7, -.6, -.5, -.4, -.3, -.2, -.1, 0., .1, .2, .3,
.4, .5, .6, .7, .8
]), unit.elementary_charge)
molecule.partial_charges = partial_charges
coords = unit.Quantity(
np.array([['0.0', '1.0', '2.0'], ['3.0', '4.0', '5.0'],
['6.0', '7.0', '8.0'], ['9.0', '10.0', '11.0'],
['12.0', '13.0', '14.0'], ['15.0', '16.0', '17.0'],
['18.0', '19.0', '20.0'], ['21.0', '22.0', '23.0'],
['24.0', '25.0', '26.0'], ['27.0', '28.0', '29.0'],
['30.0', '31.0', '32.0'], ['33.0', '34.0', '35.0'],
['36.0', '37.0', '38.0'], ['39.0', '40.0', '41.0'],
['42.0', '43.0', '44.0'], ['45.0', '46.0', '47.0'],
['48.0', '49.0', '50.0'], ['51.0', '52.0', '53.0']]),
unit.angstrom)
molecule.add_conformer(coords)
# Populate core atom property fields
molecule.atoms[2].name = 'Bob'
# Ensure one atom has its stereochemistry specified
central_carbon_stereo_specified = False
for atom in molecule.atoms:
if (atom.atomic_number == 6) and atom.stereochemistry == "S":
central_carbon_stereo_specified = True
assert central_carbon_stereo_specified
# Populate bond core property fields
fractional_bond_orders = [float(val) for val in range(18)]
for fbo, bond in zip(fractional_bond_orders, molecule.bonds):
bond.fractional_bond_order = fbo
# Do a first conversion to/from oemol
rdmol = molecule.to_rdkit()
molecule2 = Molecule.from_rdkit(rdmol)
# Test that properties survived first conversion
#assert molecule.to_dict() == molecule2.to_dict()
assert molecule.name == molecule2.name
# NOTE: This expects the same indexing scheme in the original and new molecule
central_carbon_stereo_specified = False
for atom in molecule2.atoms:
if (atom.atomic_number == 6) and atom.stereochemistry == "S":
central_carbon_stereo_specified = True
assert central_carbon_stereo_specified
for atom1, atom2 in zip(molecule.atoms, molecule2.atoms):
assert atom1.to_dict() == atom2.to_dict()
for bond1, bond2 in zip(molecule.bonds, molecule2.bonds):
assert bond1.to_dict() == bond2.to_dict()
assert (molecule._conformers[0] == molecule2._conformers[0]).all()
for pc1, pc2 in zip(molecule._partial_charges,
molecule2._partial_charges):
pc1_ul = pc1 / unit.elementary_charge
pc2_ul = pc2 / unit.elementary_charge
assert_almost_equal(pc1_ul, pc2_ul, decimal=6)
assert molecule2.to_smiles(
toolkit_registry=toolkit_wrapper) == expected_output_smiles
def test_generate_conformers(self):
"""Test RDKitToolkitWrapper generate_conformers()"""
toolkit_wrapper = RDKitToolkitWrapper()
smiles = '[H]C([H])([H])C([H])([H])[H]'
molecule = toolkit_wrapper.from_smiles(smiles)
molecule.generate_conformers()
class TestRDKitToolkitWrapper:
"""Test the RDKitToolkitWrapper"""
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_smiles(self):
"""Test RDKitToolkitWrapper to_smiles() and from_smiles()"""
toolkit_wrapper = RDKitToolkitWrapper()
# This differs from OE's expected output due to different canonicalization schemes
smiles = '[H][C]([H])([H])[C]([H])([H])[H]'
molecule = Molecule.from_smiles(smiles,
toolkit_registry=toolkit_wrapper)
smiles2 = molecule.to_smiles(toolkit_registry=toolkit_wrapper)
#print(smiles, smiles2)
assert smiles == smiles2
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
@pytest.mark.parametrize("smiles,exception_regex", [
(r"C\C(F)=C(/F)CC(C)(Cl)Br", "Undefined chiral centers"),
(r"C\C(F)=C(/F)C[C@@](C)(Cl)Br", None),
(r"CC(F)=C(F)C[C@@](C)(Cl)Br", "Bonds with undefined stereochemistry")
])
def test_smiles_missing_stereochemistry(self, smiles, exception_regex):
"""Test RDKitToolkitWrapper to_smiles() and from_smiles() when given ambiguous stereochemistry"""
toolkit_wrapper = RDKitToolkitWrapper()
if exception_regex is not None:
with pytest.raises(UndefinedStereochemistryError,
match=exception_regex):
Molecule.from_smiles(smiles, toolkit_registry=toolkit_wrapper)
Molecule.from_smiles(smiles,
toolkit_registry=toolkit_wrapper,
allow_undefined_stereo=True)
else:
Molecule.from_smiles(smiles, toolkit_registry=toolkit_wrapper)
# TODO: test_smiles_round_trip
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_smiles_add_H(self):
"""Test RDKitToolkitWrapper to_smiles() and from_smiles()"""
toolkit_wrapper = RDKitToolkitWrapper()
input_smiles = 'CC'
# This differs from OE's expected output due to different canonicalization schemes
expected_output_smiles = '[H][C]([H])([H])[C]([H])([H])[H]'
molecule = Molecule.from_smiles(input_smiles,
toolkit_registry=toolkit_wrapper)
smiles2 = molecule.to_smiles(toolkit_registry=toolkit_wrapper)
assert smiles2 == expected_output_smiles
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='OpenEye Toolkit not available')
def test_rdkit_from_smiles_hydrogens_are_explicit(self):
"""
Test to ensure that RDKitToolkitWrapper.from_smiles has the proper behavior with
respect to its hydrogens_are_explicit kwarg
"""
toolkit_wrapper = RDKitToolkitWrapper()
smiles_impl = "C#C"
with pytest.raises(
ValueError,
match=
"but RDKit toolkit interpreted SMILES 'C#C' as having implicit hydrogen"
) as excinfo:
offmol = Molecule.from_smiles(smiles_impl,
toolkit_registry=toolkit_wrapper,
hydrogens_are_explicit=True)
offmol = Molecule.from_smiles(smiles_impl,
toolkit_registry=toolkit_wrapper,
hydrogens_are_explicit=False)
assert offmol.n_atoms == 4
smiles_expl = "[H][C]#[C][H]"
offmol = Molecule.from_smiles(smiles_expl,
toolkit_registry=toolkit_wrapper,
hydrogens_are_explicit=True)
assert offmol.n_atoms == 4
# It's debatable whether this next function should pass. Strictly speaking, the hydrogens in this SMILES
# _are_ explicit, so allowing "hydrogens_are_explicit=False" through here is allowing a contradiction.
# We might rethink the name of this kwarg.
offmol = Molecule.from_smiles(smiles_expl,
toolkit_registry=toolkit_wrapper,
hydrogens_are_explicit=False)
assert offmol.n_atoms == 4
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_smiles_charged(self):
"""Test RDKitWrapper functions for reading/writing charged SMILES"""
toolkit_wrapper = RDKitToolkitWrapper()
# This differs from OE's expected output due to different canonicalization schemes
smiles = '[H][C]([H])([H])[N+]([H])([H])[H]'
molecule = Molecule.from_smiles(smiles,
toolkit_registry=toolkit_wrapper)
smiles2 = molecule.to_smiles(toolkit_registry=toolkit_wrapper)
assert smiles == smiles2
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_to_from_rdkit_core_props_filled(self):
"""Test RDKitToolkitWrapper to_rdkit() and from_rdkit() when given populated core property fields"""
toolkit_wrapper = RDKitToolkitWrapper()
# Replacing with a simple molecule with stereochemistry
input_smiles = r'C\C(F)=C(/F)C[C@@](C)(Cl)Br'
expected_output_smiles = r'[H][C]([H])([H])/[C]([F])=[C](\[F])[C]([H])([H])[C@@]([Cl])([Br])[C]([H])([H])[H]'
molecule = Molecule.from_smiles(input_smiles,
toolkit_registry=toolkit_wrapper)
assert molecule.to_smiles(
toolkit_registry=toolkit_wrapper) == expected_output_smiles
# Populate core molecule property fields
molecule.name = 'Alice'
partial_charges = unit.Quantity(
np.array([
-.9, -.8, -.7, -.6, -.5, -.4, -.3, -.2, -.1, 0., .1, .2, .3,
.4, .5, .6, .7, .8
]), unit.elementary_charge)
molecule.partial_charges = partial_charges
coords = unit.Quantity(
np.array([['0.0', '1.0', '2.0'], ['3.0', '4.0', '5.0'],
['6.0', '7.0', '8.0'], ['9.0', '10.0', '11.0'],
['12.0', '13.0', '14.0'], ['15.0', '16.0', '17.0'],
['18.0', '19.0', '20.0'], ['21.0', '22.0', '23.0'],
['24.0', '25.0', '26.0'], ['27.0', '28.0', '29.0'],
['30.0', '31.0', '32.0'], ['33.0', '34.0', '35.0'],
['36.0', '37.0', '38.0'], ['39.0', '40.0', '41.0'],
['42.0', '43.0', '44.0'], ['45.0', '46.0', '47.0'],
['48.0', '49.0', '50.0'], ['51.0', '52.0', '53.0']]),
unit.angstrom)
molecule.add_conformer(coords)
# Populate core atom property fields
molecule.atoms[2].name = 'Bob'
# Ensure one atom has its stereochemistry specified
central_carbon_stereo_specified = False
for atom in molecule.atoms:
if (atom.atomic_number == 6) and atom.stereochemistry == "S":
central_carbon_stereo_specified = True
assert central_carbon_stereo_specified
# Populate bond core property fields
fractional_bond_orders = [float(val) for val in range(18)]
for fbo, bond in zip(fractional_bond_orders, molecule.bonds):
bond.fractional_bond_order = fbo
# Do a first conversion to/from oemol
rdmol = molecule.to_rdkit()
molecule2 = Molecule.from_rdkit(rdmol)
# Test that properties survived first conversion
#assert molecule.to_dict() == molecule2.to_dict()
assert molecule.name == molecule2.name
# NOTE: This expects the same indexing scheme in the original and new molecule
central_carbon_stereo_specified = False
for atom in molecule2.atoms:
if (atom.atomic_number == 6) and atom.stereochemistry == "S":
central_carbon_stereo_specified = True
assert central_carbon_stereo_specified
for atom1, atom2 in zip(molecule.atoms, molecule2.atoms):
assert atom1.to_dict() == atom2.to_dict()
for bond1, bond2 in zip(molecule.bonds, molecule2.bonds):
assert bond1.to_dict() == bond2.to_dict()
assert (molecule._conformers[0] == molecule2._conformers[0]).all()
for pc1, pc2 in zip(molecule._partial_charges,
molecule2._partial_charges):
pc1_ul = pc1 / unit.elementary_charge
pc2_ul = pc2 / unit.elementary_charge
assert_almost_equal(pc1_ul, pc2_ul, decimal=6)
assert molecule2.to_smiles(
toolkit_registry=toolkit_wrapper) == expected_output_smiles
# TODO: This should be its own test
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_to_from_rdkit_core_props_unset(self):
"""Test RDKitToolkitWrapper to_rdkit() and from_rdkit() when given empty core property fields"""
toolkit_wrapper = RDKitToolkitWrapper()
# Replacing with a simple molecule with stereochemistry
input_smiles = r'C\C(F)=C(/F)C[C@](C)(Cl)Br'
expected_output_smiles = r'[H][C]([H])([H])/[C]([F])=[C](\[F])[C]([H])([H])[C@]([Cl])([Br])[C]([H])([H])[H]'
molecule = Molecule.from_smiles(input_smiles,
toolkit_registry=toolkit_wrapper)
assert molecule.to_smiles(
toolkit_registry=toolkit_wrapper) == expected_output_smiles
# Ensure one atom has its stereochemistry specified
central_carbon_stereo_specified = False
for atom in molecule.atoms:
if (atom.atomic_number == 6) and atom.stereochemistry == "R":
central_carbon_stereo_specified = True
assert central_carbon_stereo_specified
# Do a first conversion to/from oemol
rdmol = molecule.to_rdkit()
molecule2 = Molecule.from_rdkit(rdmol)
# Test that properties survived first conversion
assert molecule.name == molecule2.name
# NOTE: This expects the same indexing scheme in the original and new molecule
central_carbon_stereo_specified = False
for atom in molecule2.atoms:
if (atom.atomic_number == 6) and atom.stereochemistry == "R":
central_carbon_stereo_specified = True
assert central_carbon_stereo_specified
for atom1, atom2 in zip(molecule.atoms, molecule2.atoms):
assert atom1.to_dict() == atom2.to_dict()
for bond1, bond2 in zip(molecule.bonds, molecule2.bonds):
assert bond1.to_dict() == bond2.to_dict()
assert (molecule._conformers == None)
assert (molecule2._conformers == None)
for pc1, pc2 in zip(molecule._partial_charges,
molecule2._partial_charges):
pc1_ul = pc1 / unit.elementary_charge
pc2_ul = pc2 / unit.elementary_charge
assert_almost_equal(pc1_ul, pc2_ul, decimal=6)
assert molecule2.to_smiles(
toolkit_registry=toolkit_wrapper) == expected_output_smiles
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_get_sdf_coordinates(self):
"""Test RDKitToolkitWrapper for importing a single set of coordinates from a sdf file"""
toolkit_wrapper = RDKitToolkitWrapper()
filename = get_data_file_path('molecules/toluene.sdf')
molecule = Molecule.from_file(filename,
toolkit_registry=toolkit_wrapper)
assert len(molecule._conformers) == 1
assert molecule._conformers[0].shape == (15, 3)
assert_almost_equal(molecule.conformers[0][5][1] / unit.angstrom,
2.0104,
decimal=4)
# Find a multiconformer SDF file
@pytest.mark.skip
#@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(), reason='RDKit Toolkit not available')
def test_get_multiconformer_sdf_coordinates(self):
"""Test RDKitToolkitWrapper for importing a single set of coordinates from a sdf file"""
raise NotImplementedError
toolkit_wrapper = RDKitToolkitWrapper()
filename = get_data_file_path('molecules/toluene.sdf')
molecule = Molecule.from_file(filename,
toolkit_registry=toolkit_wrapper)
assert len(molecule._conformers) == 1
assert molecule._conformers[0].shape == (15, 3)
# Unskip this when we implement PDB-reading support for RDKitToolkitWrapper
@pytest.mark.skip
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_get_pdb_coordinates(self):
"""Test RDKitToolkitWrapper for importing a single set of coordinates from a pdb file"""
toolkit_wrapper = RDKitToolkitWrapper()
filename = get_data_file_path('molecules/toluene.pdb')
molecule = Molecule.from_file(filename,
toolkit_registry=toolkit_wrapper)
assert len(molecule._conformers) == 1
assert molecule._conformers[0].shape == (15, 3)
# Unskip this when we implement PDB-reading support for RDKitToolkitWrapper
@pytest.mark.skip
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_load_aromatic_pdb(self):
"""Test OpenEyeToolkitWrapper for importing molecule conformers"""
toolkit_wrapper = RDKitToolkitWrapper()
filename = get_data_file_path('molecules/toluene.pdb')
molecule = Molecule.from_file(filename,
toolkit_registry=toolkit_wrapper)
assert len(molecule._conformers) == 1
assert molecule._conformers[0].shape == (15, 3)
@pytest.mark.skipif(not RDKitToolkitWrapper.is_available(),
reason='RDKit Toolkit not available')
def test_generate_conformers(self):
"""Test RDKitToolkitWrapper generate_conformers()"""
toolkit_wrapper = RDKitToolkitWrapper()
smiles = '[H]C([H])([H])C([H])([H])[H]'
molecule = toolkit_wrapper.from_smiles(smiles)
molecule.generate_conformers()