def test_stereochem():
parser = SmilesParser()
parser.parse_smiles('N[C@](Br)(O)C')
parser_coords = [atom.coord for atom in parser.atoms][:5]
desired_coords = [[1.26597, 0.60740, -0.09729],
[-0.26307, 0.59858, -0.07141],
[-0.91282, 2.25811, 0.01409],
[-0.72365, -0.12709, 1.01313],
[-0.64392, 0.13084, -1.00380]]
assert calc_rmsd(parser_coords, desired_coords) < 0.5
parser = SmilesParser()
parser.parse_smiles('N[C@@H](Br)(O)')
parser_coords = [atom.coord
for atom in parser.atoms][:4] + [parser.atoms[6].coord]
desired_coords = [[1.26597, 0.60740, -0.09729],
[-0.26307, 0.59858, -0.07141],
[-0.72365, -0.12709, 1.01313],
[-0.91282, 2.25811, 0.01409],
[-0.64392, 0.13084, -1.00380]]
assert calc_rmsd(parser_coords, desired_coords) < 0.5
parser = SmilesParser()
parser.parse_smiles('F/C=C/F')
parser_coords = [atom.coord for atom in parser.atoms]
desired_coords = [[-4.14679, 1.36072, 0.92663],
[-3.58807, 1.44785, -0.00000],
[-2.26409, 1.31952, 0.00000],
[-1.70538, 1.40665, -0.92663],
[-4.11965, 1.64066, -0.92663],
[-1.73251, 1.12671, 0.92663]]
assert calc_rmsd(parser_coords, desired_coords) < 0.5
def test_calc_rmsd():
atoms = [
Atom('C', 0.0009, 0.0041, -0.0202),
Atom('H', -0.6577, -0.8481, -0.3214),
Atom('H', -0.4585, 0.9752, -0.3061),
Atom('H', 0.0853, -0.0253, 1.0804),
Atom('H', 1.0300, -0.1058, -0.4327)
]
atoms_rot = [
Atom('C', -0.0009, -0.0041, -0.0202),
Atom('H', 0.6577, 0.8481, -0.3214),
Atom('H', 0.4585, -0.9752, -0.3061),
Atom('H', -0.0853, 0.0253, 1.0804),
Atom('H', -1.0300, 0.1058, -0.4327)
]
coords1 = np.array([atom.coord for atom in atoms])
coords2 = np.array([atom.coord for atom in atoms_rot])
# Rotated coordinates should have almost 0 RMSD between them
assert geom.calc_rmsd(coords1, coords2) < 1E-5
# Coordinates need to have the same shape to calculate the RMSD
with pytest.raises(AssertionError):
_ = geom.calc_rmsd(coords1, coords2[1:])
assert geom.calc_heavy_atom_rmsd(atoms, atoms_rot) < 1E-5
# Permuting two hydrogens should generate a larger RMSD
atoms_rot[2], atoms_rot[3] = atoms_rot[3], atoms_rot[2]
rmsd = geom.calc_rmsd(coords1=np.array([atom.coord for atom in atoms]),
coords2=np.array([atom.coord for atom in atoms_rot]))
assert rmsd > 0.1
# While the heavy atom RMSD should remain unchanged
assert geom.calc_heavy_atom_rmsd(atoms, atoms_rot) < 1E-6
def test_chiral_rotation(tmpdir):
os.chdir(tmpdir)
chiral_ethane = Molecule(name='chiral_ethane',
charge=0,
mult=1,
atoms=[
Atom('C', -0.26307, 0.59858, -0.07141),
Atom('C', 1.26597, 0.60740, -0.09729),
Atom('Cl', -0.91282, 2.25811, 0.01409),
Atom('F', -0.72365, -0.12709, 1.01313),
Atom('H', -0.64392, 0.13084, -1.00380),
Atom('Cl', 1.93888, 1.31880, 1.39553),
Atom('H', 1.61975, 1.19877, -0.96823),
Atom('Br', 1.94229, -1.20011, -0.28203)
])
chiral_ethane.graph.nodes[0]['stereo'] = True
chiral_ethane.graph.nodes[1]['stereo'] = True
atoms = conf_gen.get_simanl_atoms(chiral_ethane)
regen = Molecule(name='regenerated_ethane', charge=0, mult=1, atoms=atoms)
regen_coords = regen.get_coordinates()
coords = chiral_ethane.get_coordinates()
# Atom indexes of the C(C)(Cl)(F)(H) chiral centre
ccclfh = [0, 1, 2, 3, 4]
# Atom indexes of the C(C)(Cl)(Br)(H) chiral centre
ccclbrh = [1, 0, 5, 7, 6]
for centre_idxs in [ccclfh, ccclbrh]:
# Ensure the fragmented centres map almost identically
# if calc_rmsd(template_coords=coords[centre_idxs], coords_to_
# fit=regen_coords[centre_idxs]) > 0.5:
# chiral_ethane.print_xyz_file(filename=os.path.join(here,
# 'chiral_ethane.xyz'))
# regen.print_xyz_file(filename=os.path.join(here, 'regen.xyz'))
# RMSD on the 5 atoms should be < 0.5 Å
assert calc_rmsd(coords1=coords[centre_idxs],
coords2=regen_coords[centre_idxs]) < 0.5
os.chdir(here)
def get_and_copy_unique_confs(xyz_filenames, only_heavy_atoms, threshold_rmsd):
"""For each xyz file generate a species and copy it to a folder if it is
unique based on an RMSD threshold"""
molecules = [
Species(name=fn.rstrip('.xyz'),
atoms=xyz_file_to_atoms(fn),
charge=0,
mult=1) for fn in xyz_filenames
]
if only_heavy_atoms:
molecules = get_molecules_no_hydrogens(molecules)
unique_mol_ids = []
for i in range(len(molecules)):
mol = molecules[i]
is_unique = True
for j in unique_mol_ids:
rmsd = calc_rmsd(coords1=mol.get_coordinates(),
coords2=molecules[j].get_coordinates())
if rmsd < threshold_rmsd:
is_unique = False
break
if is_unique:
unique_mol_ids.append(i)
print('Number of unique molecules = ', len(unique_mol_ids))
# Copy all the unique .xyz files to a new folder
if not os.path.exists(folder_name):
os.mkdir(folder_name)
for i in unique_mol_ids:
xyz_filename = xyz_filenames[i]
copyfile(xyz_filename, os.path.join(folder_name, xyz_filename))
return None