def test_ProteinAlphabetBuilder_get_resid_monomer_details(self):
path = os.path.join(self.dirname, 'alphabet.yml')
session = requests.Session()
self.assertEqual(protein.ProteinAlphabetBuilder().get_resid_monomer_details('AA0037', session)[0], 'S')
self.assertEqual(protein.ProteinAlphabetBuilder().get_resid_monomer_details('AA0037', session)[3], {'AA0016'})
identifiers = protein.ProteinAlphabetBuilder().get_resid_monomer_details('AA0037', session)[2]
identifiers_test = set([
Identifier('pdb.ligand', 'SEP'),
Identifier('mod', 'MOD:00046'),
Identifier('go', 'GO:0018105'),
Identifier('cas', '407-41-0'),
Identifier('chebi', 'CHEBI:45522'),
Identifier('resid', 'AA0037'),
])
self.assertEqual(identifiers, identifiers_test)
def test_test_ProteinAlphabetBuilder_is_terminus(self):
builder = protein.ProteinAlphabetBuilder()
mol = openbabel.OBMol()
conv = openbabel.OBConversion()
conv.SetInFormat('smiles')
conv.ReadString(mol, 'O=C[C@H](CCCCNC(=O)[C@H](CCCC[NH3+])[NH3+])[NH3+]')
self.assertTrue(builder.is_terminus(mol.GetAtom(26), mol.GetAtom(2)))
self.assertFalse(builder.is_terminus(mol.GetAtom(26), mol.GetAtom(10)))
def test_ProteinAlphabetBuilder_is_termini(self):
builder = protein.ProteinAlphabetBuilder()
mol = openbabel.OBMol()
conv = openbabel.OBConversion()
conv.SetInFormat('smiles')
conv.ReadString(mol, 'C[C@H]([NH3+])C=O')
for i_atom in range(mol.NumAtoms()):
if i_atom + 1 == 4:
self.assertTrue(builder.is_n_terminus(mol, mol.GetAtom(i_atom + 1)))
else:
self.assertFalse(builder.is_n_terminus(mol, mol.GetAtom(i_atom + 1)))
if i_atom + 1 == 8:
self.assertTrue(builder.is_c_terminus(mol, mol.GetAtom(i_atom + 1)))
else:
self.assertFalse(builder.is_c_terminus(mol, mol.GetAtom(i_atom + 1)))
self.assertFalse(builder.is_n_terminus(mol, None))
conv.ReadString(mol, 'N')
atom = mol.GetAtom(1)
self.assertFalse(builder.is_n_terminus(mol, atom))
conv.ReadString(mol, '[NH3+]O')
atom = mol.GetAtom(1)
self.assertFalse(builder.is_n_terminus(mol, atom))
conv.ReadString(mol, '[NH4+]')
atom = mol.GetAtom(1)
self.assertFalse(builder.is_n_terminus(mol, atom))
conv.ReadString(mol, 'C[NH+](C)C')
atom = mol.GetAtom(2)
self.assertFalse(builder.is_n_terminus(mol, atom))
self.assertFalse(builder.is_c_terminus(mol, None))
conv.ReadString(mol, 'C[C@H]([NH3+])C=O')
atom = mol.GetAtom(8)
atom.SetFormalCharge(1)
self.assertFalse(builder.is_c_terminus(mol, atom))
conv.ReadString(mol, 'CC(C)O')
atom = mol.GetAtom(2)
self.assertFalse(builder.is_c_terminus(mol, atom))
conv.ReadString(mol, 'CCO')
atom = mol.GetAtom(2)
self.assertFalse(builder.is_c_terminus(mol, atom))
conv.ReadString(mol, 'C=C=O')
atom = mol.GetAtom(2)
self.assertFalse(builder.is_c_terminus(mol, atom))
def test_ProteinAlphabetBuilder_get_resid_monomer_structure(self):
path = os.path.join(self.dirname, 'alphabet.yml')
structure, index_n, index_c = protein.ProteinAlphabetBuilder().get_resid_monomer_structure(
'AA0005', self.tmp_pdbfile, ph=7.4, major_tautomer=True)
# just in case check that original structure has not been modified
self.assertEqual(OpenBabelUtils.export(structure, 'smiles'),
'OC(=O)[C@@H]([NH3+])CS')
# check if correct index for N and C atoms
self.assertEqual(index_n, 6)
self.assertEqual(index_c, 2)
def test_ProteinAlphabetBuilder(self):
pdb_dir = pkg_resources.resource_filename('bpforms', os.path.join('alphabet', 'protein.pdb'))
if os.path.isdir(pdb_dir):
shutil.rmtree(pdb_dir)
path = os.path.join(self.dirname, 'alphabet.yml')
alphabet = protein.ProteinAlphabetBuilder(_max_monomers=10).run(ph=7.4, path=path)
self.assertEqual(alphabet.monomers.F.get_formula(),
EmpiricalFormula('C9H12NO') + EmpiricalFormula('O'))
self.assertTrue(os.path.isfile(path))
alphabet = Alphabet().from_yaml(path)
self.assertEqual(alphabet.monomers.F.get_formula(),
EmpiricalFormula('C9H12NO') + EmpiricalFormula('O'))
def validate_alphabet(self, alphabet):
errors = []
builder = protein.ProteinAlphabetBuilder()
for monomer in alphabet.monomers.values():
if monomer.l_bond_atoms:
atom_n = monomer.structure.GetAtom(monomer.l_bond_atoms[0].position)
if not builder.is_n_terminus(monomer.structure, atom_n):
errors.append('Monomer {} does not have a N-terminus'.format(monomer.id))
if monomer.r_bond_atoms:
atom_c = monomer.structure.GetAtom(monomer.r_bond_atoms[0].position)
if not builder.is_c_terminus(monomer.structure, atom_c, residue=False):
errors.append('Monomer {} does not have a C-terminus'.format(monomer.id))
# if monomer.l_bond_atoms and monomer.r_bond_atoms:
# if not builder.is_terminus(atom_n, atom_c):
# errors.append('Monomer {} does not have termini'.format(monomer.id))
if errors:
raise ValueError('Alphabet has invalid monomer(s):\n {}'.format('\n '.join(
errors)))