def test_example_wang_4(self) -> None:
uncapped_molecule = Molecule(
{
1:
Atom(index=1,
element='O',
valence=1,
capped=True,
coordinates=None),
2:
Atom(index=2,
element='C',
valence=3,
capped=True,
coordinates=None),
3:
Atom(index=3,
element='C',
valence=3,
capped=False,
coordinates=None),
4:
Atom(index=4,
element='C',
valence=3,
capped=False,
coordinates=None),
5:
Atom(index=5,
element='C',
valence=3,
capped=False,
coordinates=None),
6:
Atom(index=6,
element='N',
valence=2,
capped=True,
coordinates=None),
},
[
(1, 2),
(2, 3),
(3, 4),
(4, 5),
(5, 6),
(6, 2),
],
name='example_wang_4',
)
uncapped_molecule.write_graph('uncapped_molecule')
for use_ILP in (True, False):
capped_molecule = getattr(uncapped_molecule,
CAPPING_FUNCTION_NAME)(debug=None)
capped_molecule.write_graph('capped_molecule_with_{0}'.format(
'ILP' if use_ILP else 'bruteforce'))
assert capped_molecule.formula(
charge=True) == 'C4H3NO', capped_molecule.formula(charge=True)
def test_example_2(self) -> None:
uncapped_molecule = Molecule(
{
1:
Atom(index=1,
element='C',
valence=3,
capped=False,
coordinates=None),
2:
Atom(index=2,
element='O',
valence=1,
capped=False,
coordinates=None),
},
[
(1, 2),
],
name='example_2',
)
uncapped_molecule.write_graph('uncapped_molecule')
for use_ILP in (True, ):
capped_molecule = getattr(uncapped_molecule,
CAPPING_FUNCTION_NAME)(debug=None)
capped_molecule.write_graph('capped_molecule_with_{0}'.format(
'ILP' if use_ILP else 'bruteforce'))
assert capped_molecule.formula(
charge=True) == 'CH2O', capped_molecule.formula(charge=True)
def test_example_0(self) -> None:
molecule = Molecule(
[
Atom(index=1,
element='C',
valence=3,
capped=True,
coordinates=None),
Atom(index=2,
element='O',
valence=1,
capped=True,
coordinates=None),
Atom(index=3,
element='O',
valence=2,
capped=True,
coordinates=None),
Atom(index=4,
element='H',
valence=1,
capped=True,
coordinates=None),
Atom(index=5,
element='H',
valence=1,
capped=True,
coordinates=None),
],
[
(1, 2),
(1, 3),
(3, 4),
(1, 5),
],
name='methoanoic acid',
)
molecule.write_graph('raw_molecule')
molecule.assign_bond_orders_and_charges_with_ILP()
molecule.write_graph('molecule_with_electrons')
molecule.get_all_tautomers(
**OPTIONS[molecule_name] if molecule_name in
OPTIONS else {}))
else:
print(
molecule.assign_bond_orders_and_charges_with_ILP(
enforce_octet_rule=True))
print(
molecule.write_graph(molecule_name,
output_size=(int(2100 / 1.5),
int(2970 / 1.5))))
if molecule_name == 'warfarin':
print(molecule)
print()
molecule = Molecule([
Atom(index=1, element='C', valence=3, capped=True, coordinates=None),
Atom(index=2, element='C', valence=3, capped=True, coordinates=None)
], [(1, 2)])
print(molecule.get_all_tautomers())
print(molecule.write_graph('ethene', output_size=(200, 200)))
molecule = Molecule([
Atom(index=1, element='H', valence=1, capped=True, coordinates=None),
Atom(index=2, element='O', valence=1, capped=True, coordinates=None)
], [(1, 2)],
netcharge=0,
name='hydroxyl_radical')
print(molecule.assign_bond_orders_and_charges_with_ILP())
print(molecule.write_graph('', output_size=(200, 200)))
def test_example_wang_1(self) -> None:
uncapped_molecule = Molecule(
{
1:
Atom(index=1,
element='C',
valence=3,
capped=True,
coordinates=None),
2:
Atom(index=2,
element='C',
valence=3,
capped=True,
coordinates=None),
3:
Atom(index=3,
element='C',
valence=3,
capped=True,
coordinates=None),
4:
Atom(index=4,
element='C',
valence=3,
capped=True,
coordinates=None),
5:
Atom(index=5,
element='C',
valence=3,
capped=True,
coordinates=None),
6:
Atom(index=6,
element='C',
valence=3,
capped=True,
coordinates=None),
7:
Atom(index=7,
element='H',
valence=1,
capped=True,
coordinates=None),
8:
Atom(index=8,
element='H',
valence=1,
capped=True,
coordinates=None),
9:
Atom(index=9,
element='H',
valence=1,
capped=True,
coordinates=None),
10:
Atom(index=10,
element='H',
valence=1,
capped=True,
coordinates=None),
11:
Atom(index=11,
element='H',
valence=1,
capped=True,
coordinates=None),
12:
Atom(index=12,
element='O',
valence=1,
capped=True,
coordinates=None),
},
[
(1, 2),
(2, 3),
(3, 4),
(4, 5),
(5, 6),
(6, 1),
(7, 1),
(8, 2),
(9, 3),
(10, 4),
(11, 5),
(12, 6),
],
name='example_wang_1',
)
uncapped_molecule.write_graph('uncapped_molecule')
for use_ILP in (True, False):
capped_molecule = getattr(uncapped_molecule,
CAPPING_FUNCTION_NAME)(debug=None)
capped_molecule.write_graph('capped_molecule_with_{0}'.format(
'ILP' if use_ILP else 'bruteforce'))
assert capped_molecule.formula(
charge=True) == 'C6H5O 1-', capped_molecule.formula(charge=True)
def test_example_taxol_core(self) -> None:
uncapped_molecule = Molecule(
{
5:
Atom(index=5,
element="C",
valence=3,
capped=False,
coordinates=(0.746, 3.138, 0.794)),
7:
Atom(index=7,
element="O",
valence=2,
capped=True,
coordinates=(1.175, 1.853, 0.61)),
8:
Atom(index=8,
element="C",
valence=4,
capped=True,
coordinates=(1.672, 1.0, 1.641)),
9:
Atom(index=9,
element="C",
valence=4,
capped=True,
coordinates=(2.696, 1.617, 2.644)),
10:
Atom(index=10,
element="H",
valence=1,
capped=True,
coordinates=(2.871, 2.705, 2.545)),
11:
Atom(index=11,
element="H",
valence=1,
capped=True,
coordinates=(3.667, 1.113, 2.766)),
12:
Atom(index=12,
element="O",
valence=2,
capped=True,
coordinates=(1.805, 1.347, 3.756)),
13:
Atom(index=13,
element="C",
valence=4,
capped=True,
coordinates=(0.741, 0.845, 2.897)),
14:
Atom(index=14,
element="H",
valence=1,
capped=True,
coordinates=(-0.111, 1.567, 2.942)),
15:
Atom(index=15,
element="C",
valence=4,
capped=False,
coordinates=(0.262, -0.558, 3.171)),
18:
Atom(index=18,
element="C",
valence=4,
capped=False,
coordinates=(1.353, -1.632, 3.131)),
20:
Atom(index=20,
element="O",
valence=2,
capped=True,
coordinates=(1.998, -1.605, 4.383)),
21:
Atom(index=21,
element="H",
valence=1,
capped=True,
coordinates=(2.798, -2.116, 4.304)),
22:
Atom(index=22,
element="C",
valence=4,
capped=False,
coordinates=(2.315, -1.541, 1.879)),
29:
Atom(index=29,
element="C",
valence=4,
capped=True,
coordinates=(1.951, -0.352, 0.936)),
30:
Atom(index=30,
element="H",
valence=1,
capped=True,
coordinates=(0.931, -0.608, 0.548)),
31:
Atom(index=31,
element="C",
valence=4,
capped=True,
coordinates=(2.911, -0.285, -0.293)),
32:
Atom(index=32,
element="H",
valence=1,
capped=True,
coordinates=(3.702, -1.082, -0.163)),
33:
Atom(index=33,
element="O",
valence=2,
capped=False,
coordinates=(3.52, 1.006, -0.367)),
101:
Atom(index=101,
element="C",
valence=4,
capped=False,
coordinates=(0.908, -3.278, 0.308)),
47:
Atom(index=47,
element="C",
valence=4,
capped=True,
coordinates=(2.191, -0.548, -1.701)),
48:
Atom(index=48,
element="O",
valence=2,
capped=True,
coordinates=(3.132, -0.275, -2.729)),
49:
Atom(index=49,
element="H",
valence=1,
capped=True,
coordinates=(3.241, 0.668, -2.768)),
50:
Atom(index=50,
element="C",
valence=4,
capped=False,
coordinates=(0.937, 0.345, -1.879)),
53:
Atom(index=53,
element="C",
valence=4,
capped=False,
coordinates=(1.845, -2.08, -1.847)),
54:
Atom(index=54,
element="C",
valence=3,
capped=True,
coordinates=(0.733, -2.32, -0.843)),
59:
Atom(index=59,
element="C",
valence=3,
capped=False,
coordinates=(-0.388, -1.56, -0.899))
},
{
frozenset({5, 7}),
frozenset({13, 15}),
frozenset({18, 20}),
frozenset({18, 15}),
frozenset({50, 47}),
frozenset({18, 22}),
frozenset({8, 29}),
frozenset({8, 9}),
frozenset({32, 31}),
frozenset({48, 47}),
frozenset({9, 10}),
frozenset({29, 30}),
frozenset({29, 31}),
frozenset({9, 11}),
frozenset({53, 54}),
frozenset({9, 12}),
frozenset({20, 21}),
frozenset({48, 49}),
frozenset({29, 22}),
frozenset({33, 31}),
frozenset({31, 47}),
frozenset({12, 13}),
frozenset({13, 14}),
frozenset({59, 54}),
frozenset({8, 7}),
frozenset({53, 47}),
frozenset({101, 54}),
frozenset({8, 13})
},
name='example_taxol_core',
)
uncapped_molecule.write_graph('uncapped_molecule',
output_size=(1200, 1200))
for use_ILP in (True, ):
capped_molecule = getattr(uncapped_molecule,
CAPPING_FUNCTION_NAME)(debug=None)
capped_molecule.write_graph('capped_molecule_with_{0}'.format(
'ILP' if use_ILP else 'bruteforce'),
output_size=(1200, 1200))
assert capped_molecule.formula(
charge=True) == 'C16H26O5', capped_molecule.formula(charge=True)
CAN_FAIL, CAN_NOT_FAIL = True, False
USE_OCTET_RULE, DO_NOT_USE_OCTET_RULE = True, False
ALL_EXAMPLES = {
Molecule(
[
Atom(index=1, element='C', valence=4, capped=True, coordinates=None),
Atom(index=2, element='C', valence=3, capped=True, coordinates=None),
Atom(index=3, element='H', valence=None, capped=False, coordinates=None),
Atom(index=4, element='H', valence=None, capped=False, coordinates=None),
Atom(index=5, element='H', valence=None, capped=False, coordinates=None),
Atom(index=6, element='N', valence=None, capped=False, coordinates=None),
Atom(index=7, element='C', valence=None, capped=False, coordinates=None),
],
[
(1, 2),
(1, 3),
(1, 4),
(1, 5),
(2, 6),
(2, 7),
],
name='H,H,H_C_C_N,C',
): (0, 0, 'C3H7N', USE_OCTET_RULE, CAN_NOT_FAIL),
Molecule(
[
Atom(index=1, element='C', valence=4, capped=True, coordinates=None),
Atom(index=2, element='C', valence=4, capped=True, coordinates=None),
Atom(index=3, element='H', valence=None, capped=False, coordinates=None),
Atom(index=4, element='H', valence=None, capped=False, coordinates=None),
def uncapped_molecule_for_dihedral_fragment(dihedral_fragment: Fragment,
debug: bool = False
) -> Uncapped_Molecule:
if dihedral_fragment.count('|') == 3:
neighbours_1, atom_2, atom_3, neighbours_4 = dihedral_fragment.split(
'|')
cycles = []
neighbours_1, neighbours_4 = neighbours_1.split(
','), neighbours_4.split(',')
elif dihedral_fragment.count('|') == 4:
neighbours_1, atom_2, atom_3, neighbours_4, cycles = dihedral_fragment.split(
'|')
neighbours_1, neighbours_4, cycles = neighbours_1.split(
','), neighbours_4.split(','), cycles.split(',')
else:
raise Exception(
'Invalid dihedral_fragment: "{0}"'.format(dihedral_fragment))
ids = [
n
for (n, _) in enumerate(neighbours_1 + [atom_2, atom_3] + neighbours_4,
start=1)
]
neighbours_id_1, atom_id_2, atom_id_3, neighbours_id_4 = ids[0:len(
neighbours_1)], ids[len(neighbours_1)], ids[len(neighbours_1) +
1], ids[len(neighbours_1) +
2:]
CENTRAL_BOND = (atom_id_2, atom_id_3)
elements = dict(
list(
zip(
ids,
[
element_valence_for_atom(neighbour)[0]
for neighbour in neighbours_1
] + [atom_2, atom_3] + [
element_valence_for_atom(neighbour)[0]
for neighbour in neighbours_4
],
)), )
valences = dict(
list(
zip(
ids,
[
element_valence_for_atom(neighbour)[1]
for neighbour in neighbours_1
] + [len(neighbours_1) + 1,
len(neighbours_4) + 1] + [
element_valence_for_atom(neighbour)[1]
for neighbour in neighbours_4
],
)), )
bonds = ([(neighbour_id, atom_id_2) for neighbour_id in neighbours_id_1] +
[CENTRAL_BOND] + [(atom_id_3, neighbour_id)
for neighbour_id in neighbours_id_4])
def coordinates_for_atom_id(atom_id: int,
d: float = 1.5) -> Tuple[float, float, float]:
if atom_id == atom_id_2:
return (-d / 2, 0, 0)
elif atom_id == atom_id_3:
return (d / 2, 0, 0)
else:
e = 0.5 * d
f = sqrt(d**2 - e**2)
assert e**2 + f**2 == d**2, (e**2 + f**2, d**2)
if atom_id in neighbours_id_1:
left_theta = 2 * pi / len(
neighbours_id_1) * neighbours_id_1.index(atom_id)
return (-d / 2 - e, -f * cos(left_theta), f * sin(left_theta))
elif atom_id in neighbours_id_4:
right_theta = 2 * pi / len(
neighbours_id_4) * neighbours_id_4.index(atom_id)
return (d / 2 + e, f * cos(right_theta), f * sin(right_theta))
else:
raise Exception('Impossible id: {0}'.format(atom_id))
molecule = Molecule(
dict(
list(
zip(
ids,
[
Atom(
index=atom_id,
element=elements[atom_id],
valence=valences[atom_id],
capped=atom_id
not in (neighbours_id_1 + neighbours_id_4),
coordinates=coordinates_for_atom_id(atom_id),
) for atom_id in ids
],
))),
bonds,
name=dihedral_fragment.replace('|', '_'),
)
if debug:
print(molecule)
for (i, n, j) in map(lambda cycle: map(int, cycle), cycles):
i_id, j_id = neighbours_id_1[i], neighbours_id_4[j]
if n == 0:
# i and j are actually the same atoms
del molecule.atoms[j_id]
replace_j_by_i = lambda x: i_id if x == j_id else x
molecule.bonds = {
frozenset(map(replace_j_by_i, bond))
for bond in molecule.bonds
}
else:
NEW_ATOM_ID = -1
NEW_ATOM = Atom(
index=
NEW_ATOM_ID, # This will get overwritten by Molecule.add_atom
element='C',
valence=NO_VALENCE,
capped=False,
coordinates=None,
)
atom_chain_id = [i_id] + [
molecule.add_atom(NEW_ATOM) for i in range(n - 1)
] + [j_id]
new_bonds = zip(atom_chain_id[:-1], atom_chain_id[1:])
molecule.add_bonds(new_bonds)
if debug:
print(molecule)
return molecule