def test_stereoisomers_produce_equal_fingerprints_nonstereo(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
mol1 = mol_from_sdf(ENANT1_SDF_FILE)
mol2 = mol_from_sdf(ENANT2_SDF_FILE)
level = 5
fpr = fprinter.Fingerprinter(level=level, stereo=False,
radius_multiplier=1.718,
remove_duplicate_substructs=True)
fpr.run(conf=0, mol=mol1)
fp1 = fpr.get_fingerprint_at_level(level)
fpr.run(conf=0, mol=mol2)
fp2 = fpr.get_fingerprint_at_level(level)
self.assertEqual(fp1, fp2)
def test_reordering_conformers_produces_same_fprints(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
import random
rand_sdf_files = glob.glob(os.path.join(RAND_SDF_DIR, "*.sdf*"))
mol = mol_from_sdf(rand_sdf_files[0])
level = 5
fpr = fprinter.Fingerprinter(
level=level,
stereo=False,
radius_multiplier=1.718,
remove_duplicate_substructs=True,
)
conf_ids1 = [x.GetId() for x in mol.GetConformers()]
fprints1 = {}
for conf_id in conf_ids1:
fpr.run(conf_id, mol)
fprints1[conf_id] = fpr.get_fingerprint_at_level(level)
conf_ids2 = list(conf_ids1)
random.shuffle(conf_ids2)
fprints2 = {}
for conf_id in conf_ids2:
fpr.run(conf_id, mol)
fprints2[conf_id] = fpr.get_fingerprint_at_level(level)
self.assertEqual(fprints1, fprints2)
def test_generates_correct_disconnected_shells_level2(self):
from e3fp.fingerprint.fprinter import ShellsGenerator
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = list(range(3))
for atom in atoms:
conf.SetAtomPosition(atom, [0, 0, .45 * atom])
expected_shells_dict1 = {
0: Shell(0, {1}),
1: Shell(1, {0, 2}),
2: Shell(2, {1})
}
expected_shells_dict2 = {
0: Shell(0, {expected_shells_dict1[1], expected_shells_dict1[2]}),
1: Shell(1, {expected_shells_dict1[0], expected_shells_dict1[2]}),
2: Shell(2, {expected_shells_dict1[0], expected_shells_dict1[1]})
}
shells_gen = ShellsGenerator(conf,
atoms,
radius_multiplier=0.5,
include_disconnected=True)
for i in range(3):
shells_dict = next(shells_gen)
self.assertDictEqual(shells_dict, expected_shells_dict2)
def test_connected_substructs_converge(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = list(range(3))
bonds_dict = {0: {1, 2}, 1: {0}, 2: {0}}
for atom in atoms:
conf.SetAtomPosition(atom, [0, 0, .45 * atom])
with mock.patch('e3fp.fingerprint.fprinter.bound_atoms_from_mol',
return_value=bonds_dict):
shells_gen = fprinter.ShellsGenerator(conf,
atoms,
radius_multiplier=0.5,
include_disconnected=False)
for i in range(4):
shells_dict = next(shells_gen)
substructs_dict = {
k: v.substruct
for k, v in shells_dict.items()
}
next_shells_dict = next(shells_gen)
next_substructs_dict = {
k: v.substruct
for k, v in next_shells_dict.items()
}
self.assertDictEqual(substructs_dict, next_substructs_dict)
def test_generates_correct_connected_shells_level2(self):
from e3fp.fingerprint import fprinter
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = list(range(3))
bonds_dict = {0: {1, 2}, 1: {0}, 2: {0}}
for atom in atoms:
conf.SetAtomPosition(atom, [0, 0, .45 * atom])
expected_shells_dict1 = {
0: Shell(0, {1}),
1: Shell(1, {0}),
2: Shell(2, {})
}
expected_shells_dict2 = {
0: Shell(0, {expected_shells_dict1[1], expected_shells_dict1[2]}),
1: Shell(1, {expected_shells_dict1[0]}),
2: Shell(2, {expected_shells_dict1[0]})
}
with mock.patch('e3fp.fingerprint.fprinter.bound_atoms_from_mol',
return_value=bonds_dict):
shells_gen = fprinter.ShellsGenerator(conf,
atoms,
radius_multiplier=0.5,
include_disconnected=False)
for i in range(3):
shells_dict = next(shells_gen)
self.assertDictEqual(shells_dict, expected_shells_dict2)
def test_create_shell_no_shell(self):
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
center_atom = atoms[0]
Shell(center_atom)
def test_remove_dupe_substructs_makes_same_substruts_diff_shells(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
level = 2
conf = mol.GetConformers()[0]
fpr = fprinter.Fingerprinter(level=level,
bits=1024,
stereo=True,
radius_multiplier=1.718,
remove_duplicate_substructs=True)
fpr.run(conf, mol)
shells_no_dupes = set(fpr.level_shells[fpr.current_level])
substructs_no_dupes = set([x.substruct for x in shells_no_dupes])
fpr = fprinter.Fingerprinter(level=level,
bits=1024,
stereo=True,
radius_multiplier=1.718,
remove_duplicate_substructs=False)
fpr.run(conf, mol)
shells_with_dupes = set(fpr.level_shells[fpr.current_level])
substructs_with_dupes = set([x.substruct for x in shells_with_dupes])
self.assertEqual(substructs_no_dupes, substructs_with_dupes)
self.assertNotEqual(shells_no_dupes, shells_with_dupes)
def test_rdkit_invariants(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atom = mol.GetAtomWithIdx(2)
invars = fprinter.rdkit_invariants_from_atom(atom)
self.assertListEqual(list(invars), [6, 3, 1, 0, 0, 1])
def test_shells_diff_center_same_atoms_nonequal(self):
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
shell1 = Shell(atoms[0], atoms[2:])
shell2 = Shell(atoms[1], atoms[2:])
self.assertNotEqual(shell1, shell2)
def test_create_shell_with_same_center_fails(self):
from e3fp.fingerprint.structs import Shell, FormatError
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
center_atom = atoms[0]
with self.assertRaises(FormatError):
Shell(center_atom, atoms)
def test_same_shells_hash_to_same_value(self):
from e3fp.fingerprint.structs import Substruct
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
center_atom = atoms[0]
substruct = Substruct(center_atom, atoms[1:])
self.assertEqual(hash(substruct), hash(substruct))
def test_creation_with_atoms_or_ids_equivalent(self):
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
atom_ids = [x.GetIdx() for x in atoms]
self.assertEqual(Shell(atoms[0], atoms[1:]),
Shell(atom_ids[0], atom_ids[1:]))
def test_substructs_same_center_diff_atoms_nonequal(self):
from e3fp.fingerprint.structs import Substruct
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
substruct1 = Substruct(atoms[0], atoms[1:])
substruct2 = Substruct(atoms[0], atoms[2:])
self.assertNotEqual(substruct1, substruct2)
def test_center_atom_auto_added_to_atoms(self):
from e3fp.fingerprint.structs import Substruct
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
center_atom = atoms[0]
substruct = Substruct(center_atom, atoms[1:])
self.assertIn(center_atom.GetIdx(), substruct.atoms)
def test_substruct_creation_from_shell(self):
from e3fp.fingerprint.structs import Shell, Substruct
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
shell = Shell(atoms[0], atoms[1:])
substruct = Substruct.from_shell(shell)
self.assertEqual(shell.substruct, substruct)
def test_shells_generator_creation_success(self):
from e3fp.fingerprint.fprinter import ShellsGenerator
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = [x.GetIdx() for x in mol.GetAtoms()]
ShellsGenerator(conf, atoms, radius_multiplier=0.5,
include_disconnected=True)
def test_shell_creation_from_substruct_without_center_fails(self):
from e3fp.fingerprint.structs import Shell, Substruct, FormatError
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
substruct = Substruct(None, atoms[:2])
with self.assertRaises(FormatError):
Shell.from_substruct(substruct)
def main(sdf_dir, mol_file, num_confs=10000,
out_conf_file="random_conformers.txt",
out_sdf_file="random_conformers.sdf.bz2",
out_mol_file="random_conformers.csv.bz2"):
confs = set()
if os.path.isfile(out_mol_file):
logging.info("Loading existing random molecules.")
_, conf_mol_list_dict, _ = molecules_to_lists_dicts(out_mol_file,
merge_proto=False)
for proto_name in conf_mol_list_dict:
for _, conf_name in conf_mol_list_dict[proto_name]:
confs.add(split_conf_name(conf_name))
else:
logging.info("Loading molecules file.")
smiles_dict, mol_list_dict, fp_type = molecules_to_lists_dicts(
mol_file, merge_proto=False)
mol_name_to_proto_names = {}
for proto_name in mol_list_dict:
mol_name, _ = split_conf_name(proto_name)
mol_name_to_proto_names.setdefault(mol_name, []).append(proto_name)
conf_mol_list_dict = {}
logging.info("Picking random molecules.")
while len(confs) < num_confs:
mol_name = random.choice(mol_name_to_proto_names.keys())
proto_name = random.choice(mol_name_to_proto_names[mol_name])
_, conf_name = random.choice(mol_list_dict[proto_name])
conf = split_conf_name(conf_name)
confs.add(conf)
conf_mol_list_dict.setdefault(proto_name, set()).add(
mol_list_dict[proto_name][conf[2]])
if len(confs) % 100 == 0:
logging.info(len(confs))
conf_mol_list_dict = {k: sorted(v) for k, v
in conf_mol_list_dict.items()}
lists_dicts_to_molecules(out_mol_file, smiles_dict, conf_mol_list_dict,
fp_type)
confs = sorted(confs)
logging.info("Writing mol names to file.")
with open(out_conf_file, "w") as f:
for conf in confs:
f.write("{}\n".format(join_conf_name(*conf)))
logging.info("Saving mols to SDF file.")
with smart_open(out_sdf_file, "wb") as f:
writer = rdkit.Chem.SDWriter(f)
for j, conf in enumerate(confs):
mol_name, proto_id, conf_id = conf
sdf_file = glob.glob(os.path.join(
sdf_dir, "{}.sdf*".format(
join_conf_name(mol_name, proto_id))))[0]
mol = mol_from_sdf(sdf_file, conf_num=conf_id + 1)
name = join_conf_name(*conf)
mol.SetProp("_Name", name)
writer.write(mol, confId=conf_id)
if j > 0 and j % 10 == 0:
logging.info(j)
writer.close()
def test_substructs_same_center_same_atoms_equal(self):
from e3fp.fingerprint.structs import Substruct
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
center_atom = atoms[0]
substruct1 = Substruct(center_atom, atoms)
substruct2 = Substruct(center_atom, atoms)
self.assertEqual(substruct1, substruct2)
def test_atoms_converted_to_shells(self):
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
center_atom = atoms[0]
shell = Shell(center_atom, atoms[1:])
for s in shell.shells:
self.assertIsInstance(s, Shell)
def test_quick(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
level = 5
conf = mol.GetConformers()[0]
fpr = fprinter.Fingerprinter(level=level, bits=1024, stereo=True,
radius_multiplier=1.718)
fpr.run(conf, mol)
def test_creation_with_atoms_or_shells_equal(self):
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
shells = list(map(Shell, atoms))
center_atom = atoms[0]
shell1 = Shell(center_atom, atoms[1:])
shell2 = Shell(center_atom, shells[1:])
self.assertEqual(shell1, shell2)
def test_recursive_atom_shells_correct(self):
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
atoms = list(mol.GetAtoms())
shell1 = Shell(atoms[5], atoms[6:8])
shell2 = Shell(atoms[2], atoms[3:5])
shell = Shell(atoms[0], (shell1, shell2))
self.assertEqual(shell.atoms,
{x.GetIdx()
for x in (atoms[0], atoms[2], atoms[5])})
def test_generates_correct_disconnected_shells_level0(self):
from e3fp.fingerprint.fprinter import ShellsGenerator
from e3fp.fingerprint.structs import Shell
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = list(range(3))
expected_shells_dict = {0: Shell(0), 1: Shell(1), 2: Shell(2)}
shells_gen = ShellsGenerator(conf, atoms)
shells_dict = next(shells_gen)
self.assertDictEqual(shells_dict, expected_shells_dict)
def test_connected_match_atoms_rad0_correct(self):
from e3fp.fingerprint.fprinter import ShellsGenerator
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = list(range(3))
shells_gen = ShellsGenerator(conf, atoms, radius_multiplier=0.5,
include_disconnected=True)
match_atoms = shells_gen.get_match_atoms(0.)
expect_match_atoms = {k: set() for k in atoms}
self.assertDictEqual(match_atoms, expect_match_atoms)
def test_shells_generator_next_works_correctly(self):
from e3fp.fingerprint.fprinter import ShellsGenerator
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = [x.GetIdx() for x in mol.GetAtoms()]
shells_gen1 = ShellsGenerator(conf, atoms, radius_multiplier=0.5,
include_disconnected=True)
shells_gen2 = ShellsGenerator(conf, atoms, radius_multiplier=0.5,
include_disconnected=True)
self.assertDictEqual(next(shells_gen1), next(shells_gen2))
def test_atom_coords_calculated_correctly(self):
from e3fp.fingerprint.fprinter import coords_from_atoms
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = [x.GetIdx() for x in mol.GetAtoms()]
for atom in atoms:
conf.SetAtomPosition(atom, [0, 0, 0])
atom_coords = coords_from_atoms(atoms, conf)
expected_coords = dict(
list(zip(atoms, np.zeros((len(atoms), 3), dtype=np.float))))
np.testing.assert_equal(atom_coords, expected_coords)
def fprints_dict_from_sdf(sdf_file, **kwargs):
"""Build fingerprints dict for conformers encoded in an SDF file.
See `fprints_dict_from_mol` for description of arguments.
"""
try:
mol = mol_from_sdf(sdf_file)
except:
logging.error("Error retrieving mol from {!s}.".format(sdf_file))
return False
fprints_dict = fprints_dict_from_mol(mol, **kwargs)
return fprints_dict
def test_initial_identifiers_assigned_correctly(self):
from e3fp.fingerprint import fprinter
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
level = 0
conf = mol.GetConformers()[0]
fpr = fprinter.Fingerprinter(level=level, bits=1024, stereo=True,
radius_multiplier=1.718)
fpr.run(conf, mol)
fprint = fpr.get_fingerprint_at_level(0)
expect_ident = set([48, 124, 185, 484, 617, 674])
self.assertEqual(set(fprint.indices), expect_ident)
def test_connected_match_atoms_rad1_correct2(self):
from e3fp.fingerprint.fprinter import ShellsGenerator
from e3fp.conformer.util import mol_from_sdf
mol = mol_from_sdf(PLANAR_SDF_FILE)
conf = mol.GetConformers()[0]
atoms = list(range(3))
for atom in atoms:
conf.SetAtomPosition(atom, [0, 0, atom*.75])
shells_gen = ShellsGenerator(conf, atoms, radius_multiplier=0.5,
include_disconnected=True)
match_atoms = shells_gen.get_match_atoms(1.)
expect_match_atoms = {0: {1}, 1: {0, 2}, 2: {1}}
self.assertDictEqual(match_atoms, expect_match_atoms)