def test_get_formula_and_connectivity(self):
glc6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h2-10H,1H2,(H2,11,12,13)/t2-,3-,4+,5-,6-/m1/s1'
)
self.assertEqual(glc6p.get_formula_and_connectivity(),
'C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8')
water = molecule_util.InchiMolecule('InChI=1S/H2O/h1H2')
self.assertEqual(water.get_formula_and_connectivity(), 'H2O')
def test_is_protonation_isomer(self):
a = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h2')
b = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h3')
c = molecule_util.InchiMolecule('InChI=1S/C6H13O9P/c7/h4')
self.assertTrue(a.is_protonation_isomer(a))
self.assertTrue(b.is_protonation_isomer(b))
self.assertFalse(a.is_protonation_isomer(c))
def test_is_tautomer(self):
a = molecule_util.InchiMolecule(
'InChI=1S/C5H5N5O/c6-5-9-3-2(4(11)10-5)7-1-8-3/h1H,(H4,6,7,8,9,10,11)/t1'
)
b = molecule_util.InchiMolecule(
'InChI=1S/C5H5N5O/c6-5-9-3-2(4(11)10-5)7-1-8-3/h1H,(H4,6,7,8,9,10,11)/t2'
)
c = molecule_util.InchiMolecule(
'InChI=1S/C5H5N5O/c6/h1H,(H4,6,7,8,9,10,11)/t2')
self.assertTrue(a.is_tautomer(a))
self.assertTrue(a.is_tautomer(a))
self.assertFalse(a.is_tautomer(c))
def test_is_stereoisomer(self):
glc6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h2-10H,1H2,(H2,11,12,13)/t2-,3-,4+,5-,6-/m1/s1'
)
gal6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h2-10H,1H2,(H2,11,12,13)/t2-,3+,4+,5-,6?/m1/s1'
)
fru6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-1-3(8)5(10)6(11)4(9)2-15-16(12,13)14/h4-7,9-11H,1-2H2,(H2,12,13,14)/t4-,5-,6-/m1/s1'
)
self.assertTrue(glc6p.is_stereoisomer(glc6p))
self.assertTrue(glc6p.is_stereoisomer(gal6p))
self.assertFalse(glc6p.is_stereoisomer(fru6p))
def test_is_equal(self):
a = molecule_util.InchiMolecule('InChI=1S/BrH/h1H/p-1')
c = molecule_util.InchiMolecule('InChI=1S/BrH/h1H')
self.assertTrue(a.is_equal(a))
self.assertFalse(a.is_equal(c))
glc6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h2-10H,1H2,(H2,11,12,13)/t2-,3-,4+,5-,6-/m1/s1'
)
gal6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-3-2(1-14-16(11,12)13)15-6(10)5(9)4(3)8/h2-10H,1H2,(H2,11,12,13)/t2-,3+,4+,5-,6?/m1/s1'
)
fru6p = molecule_util.InchiMolecule(
'InChI=1S/C6H13O9P/c7-1-3(8)5(10)6(11)4(9)2-15-16(12,13)14/h4-7,9-11H,1-2H2,(H2,12,13,14)/t4-,5-,6-/m1/s1'
)
self.assertTrue(glc6p.is_equal(glc6p, check_stereochemistry=False))
self.assertTrue(glc6p.is_equal(gal6p, check_stereochemistry=False))
self.assertFalse(glc6p.is_equal(fru6p, check_stereochemistry=False))
def test(self):
inchi = 'InChI=1S/C3H4O3/c1-2(4)3(5)6/h1H3,(H,5,6)'
layers = molecule_util.InchiMolecule(inchi)
self.assertEqual(
layers.__dict__, {
'formula': 'C3H4O3',
'connections': '1-2(4)3(5)6',
'hydrogens': '1H3,(H,5,6)',
'protons': '',
'charge': '',
'double_bonds': '',
'stereochemistry': '',
'stereochemistry_parity': '',
'stereochemistry_type': '',
'isotopes': '',
'fixed_hydrogens': '',
'reconnected_metals': '',
})
self.assertEqual(str(layers), inchi)
inchi = 'InChI=1S/C3H4O3/c1-2(4)3(5)6'
layers = molecule_util.InchiMolecule(inchi)
self.assertEqual(
layers.__dict__, {
'formula': 'C3H4O3',
'connections': '1-2(4)3(5)6',
'hydrogens': '',
'protons': '',
'charge': '',
'double_bonds': '',
'stereochemistry': '',
'stereochemistry_parity': '',
'stereochemistry_type': '',
'isotopes': '',
'fixed_hydrogens': '',
'reconnected_metals': '',
})
self.assertEqual(str(layers), inchi)
inchi = 'InChI=1S/Ni/q+2'
self.assertEqual(str(molecule_util.InchiMolecule(inchi)), inchi)
inchi = 'InChI=1S/BrH/h1H/p-1'
self.assertEqual(str(molecule_util.InchiMolecule(inchi)), inchi)
inchi = 'InChI=1S/p+1'
self.assertEqual(str(molecule_util.InchiMolecule(inchi)), inchi)
inchi = 'InChI=1S/4O.V/q;3*-1;'
self.assertEqual(str(molecule_util.InchiMolecule(inchi)), inchi)
inchi_1 = 'InChI=1/C6H12O6/c7-1-3-4(9)5(10)6(11,2-8)12-3/h3-5,7-11H,1-2H2/t3-,4-,5+,6u/m0/s1'
inchi_2 = 'InChI=1S/C6H12O6/c7-1-3-4(9)5(10)6(11,2-8)12-3/h3-5,7-11H,1-2H2/t3-,4-,5+,6u/m0/s1'
self.assertEqual(str(molecule_util.InchiMolecule(inchi_1)), inchi_2)
def to_inchi(self, only_formula_and_connectivity=False):
""" Get the structure in InChi format
Args:
only_formula_and_connectivity (:obj:`bool`): if :obj:`True`, return only the
formula and connectivity layers
Returns:
:obj:`str`: structure in InChi format or just the formula and connectivity layers
if :obj:`only_formula_and_connectivity` is :obj:`True`
"""
inchi = molecule_util.Molecule(structure=self.structure).to_inchi()
if only_formula_and_connectivity:
return molecule_util.InchiMolecule(
inchi).get_formula_and_connectivity()
else:
return inchi
def get_concentration_by_structure(self, inchi, only_formula_and_connectivity=True, select = models.Concentration):
"""
Args:
inchi (:obj:`str`): inchi structure to find concentrations
Returns:
:obj:`list`: List of models.Concentration Objects
"""
q = self.data_source.session.query(select).join((models.Metabolite, select.metabolite)).\
join((models.Structure, models.Metabolite.structure))
if only_formula_and_connectivity:
formula_and_connectivity = molecule_util.InchiMolecule(inchi).get_formula_and_connectivity()
condition = models.Structure._structure_formula_connectivity == formula_and_connectivity
else:
condition = models.Structure._value_inchi == inchi
return q.filter(condition).all()
def get_metabolites_by_structure(self, inchi, only_formula_and_connectivity=False, select=models.Metabolite):
""" Get metabolites with the same structure. Optionally, get metabolites which only have
the same core empirical formula and core atom connecticity (i.e. same InChI formula
and connectivity layers).
Args:
inchi (:obj:`str`): molecule structure in InChI format
only_formula_and_connectivity (:obj:`bool`, optional): if :obj:`True`, get metabolites which only have
the same core empirical formula and core atom connecticity. if :obj:`False`, get metabolites with the
identical structure.
Returns:
:obj:`sqlalchemy.orm.query.Query`: query for matching metabolites
"""
q = self.data_source.session.query(select).join((models.Structure, models.Metabolite.structure))
if only_formula_and_connectivity:
formula_and_connectivity = molecule_util.InchiMolecule(inchi).get_formula_and_connectivity()
condition = models.Structure._structure_formula_connectivity == formula_and_connectivity
else:
condition = models.Structure._value_inchi == inchi
return q.filter(condition)
def load_content(self):
""" Download the content of ECMDB and store it to a local sqlite database. """
db_session = self.session
req_session = self.requests_session
# download content from server
if self.verbose:
print('Downloading compound IDs ...')
response = req_session.get(self.DOWNLOAD_INDEX_URL)
response.raise_for_status()
if self.verbose:
print(' done')
# unzip and parse content
if self.verbose:
print('Parsing compound IDs ...')
with zipfile.ZipFile(io.BytesIO(response.content), 'r') as zip_file:
with zip_file.open('ecmdb.json', 'r') as json_file:
entries = json.load(json_file)
if self.verbose:
print(' found {} compounds'.format(len(entries)))
# sort entires
entries.sort(key=lambda e: e['m2m_id'])
# limit number of processed entries
if len(entries) > self.max_entries:
entries = entries[0:self.max_entries]
# load content into sqlite database
if self.verbose:
print('Downloading {} compounds ...'.format(len(entries)))
xml_parser = jxmlease.Parser()
for i_entry, entry in enumerate(entries):
if self.verbose and (i_entry % 10 == 0):
print(' Downloading compound {} of {}'.format(
i_entry + 1, len(entries)))
# get details
response = req_session.get(
self.DOWNLOAD_COMPOUND_URL.format(entry['m2m_id']))
try:
response.raise_for_status()
except requests.exceptions.HTTPError:
warnings.warn(
'Unable to download data for compound {}'.format(
entry['m2m_id']), data_source.DataSourceWarning)
continue
entry_details = xml_parser(response.text)['compound']
compound = self.get_or_create_object(Compound,
id=self.get_node_text(
entry_details['m2m_id']))
if 'name' in entry_details:
compound.name = self.get_node_text(entry_details['name'])
if 'description' in entry_details:
compound.description = self.get_node_text(
entry_details['description'])
compound.structure = self.get_node_text(entry_details['inchi'])
if not compound.structure:
response2 = req_session.get(
self.DOWNLOAD_COMPOUND_STRUCTURE_URL.format(
entry['m2m_id']))
response2.raise_for_status()
compound.structure = response2.text
compound.comment = entry['comment']
compound.created = dateutil.parser.parse(
self.get_node_text(
entry_details['creation_date'])).replace(tzinfo=None)
compound.updated = dateutil.parser.parse(
self.get_node_text(
entry_details['update_date'])).replace(tzinfo=None)
# calculate core InChI layers to facilitate searching
try:
compound._structure_formula_connectivity = molecule_util.InchiMolecule(compound.structure) \
.get_formula_and_connectivity()
except ValueError:
warnings.warn(
'Unable to encode structure for {} in InChI'.format(
entry['m2m_id']), data_source.DataSourceWarning)
compound._structure_formula_connectivity = None
# synonyms
compound.synonyms = []
if 'iupac_name' in entry_details:
node = entry_details['iupac_name']
name = self.get_node_text(node)
compound.synonyms.append(
self.get_or_create_object(Synonym, name=name))
if 'traditional_iupac' in entry_details:
node = entry_details['traditional_iupac']
name = self.get_node_text(node)
compound.synonyms.append(
self.get_or_create_object(Synonym, name=name))
parent_node = entry_details['synonyms']
if 'synonym' in parent_node:
nodes = self.get_node_children(parent_node, 'synonym')
for node in nodes:
name = self.get_node_text(node)
compound.synonyms.append(
self.get_or_create_object(Synonym, name=name))
# locations
compound.compartments = []
parent_node = entry_details['cellular_locations']
if 'cellular_location' in parent_node:
nodes = self.get_node_children(parent_node,
'cellular_location')
for node in nodes:
name = self.get_node_text(node)
compound.compartments.append(
self.get_or_create_object(Compartment, name=name))
# todo (enhancement): parse experimental properties
# * state
# * melting_point
# * water_solubility
# * logp_hydrophobicity
# concentrations
compound.concentrations = []
parent_node = entry_details['concentrations']
if 'concentration' in parent_node:
values = self.get_node_children(parent_node, 'concentration')
errors = self.get_node_children(parent_node, 'error')
units = self.get_node_children(parent_node,
'concentration_units')
strains = self.get_node_children(parent_node, 'strain')
statuses = self.get_node_children(parent_node, 'growth_status')
medias = self.get_node_children(parent_node, 'growth_media')
temperatures = self.get_node_children(parent_node,
'temperature')
systems = self.get_node_children(parent_node, 'growth_system')
references = self.get_node_children(parent_node, 'reference')
for i_conc in range(len(values)):
value = float(self.get_node_text(values[i_conc]))
error = float(self.get_node_text(errors[i_conc]) or 'nan')
unit = self.get_node_text(units[i_conc])
if unit == 'uM':
pass
else:
raise ValueError('Unsupport units: {}'.format(unit))
if temperatures[i_conc]:
temperature, unit = self.get_node_text(
temperatures[i_conc]).split(' ')
temperature = float(temperature)
if unit != 'oC':
raise ValueError(
'Unsupport units: {}'.format(unit))
else:
temperature = None
concentration = Concentration(
value=value,
error=error,
strain=self.get_node_text(strains[i_conc]) or None,
growth_status=self.get_node_text(statuses[i_conc])
or None,
media=self.get_node_text(medias[i_conc]) or None,
temperature=temperature,
growth_system=self.get_node_text(systems[i_conc])
or None,
)
db_session.add(concentration)
if 'pubmed_id' in references[i_conc]:
pmid_nodes = self.get_node_children(
references[i_conc], 'pubmed_id')
for node in pmid_nodes:
id = self.get_node_text(node)
concentration.references.append(
self.get_or_create_object(Resource,
namespace='pubmed',
id=id))
compound.concentrations.append(concentration)
# cross references
compound.cross_references = []
id = self.get_node_text(entry_details['biocyc_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='biocyc',
id=id))
id = self.get_node_text(entry_details['cas_registry_number'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource, namespace='cas',
id=id))
id = self.get_node_text(entry_details['chebi_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='chebi',
id='CHEBI:' + id))
id = self.get_node_text(entry_details['chemspider_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='chemspider',
id=id))
id = self.get_node_text(entry_details['foodb_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='foodb.compound',
id=id))
id = self.get_node_text(entry_details['het_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='ligandexpo',
id=id))
id = self.get_node_text(entry_details['hmdb_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='hmdb',
id=id))
id = self.get_node_text(entry_details['kegg_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='kegg.compound',
id=id))
id = self.get_node_text(entry_details['msds_url'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='msds.url',
id=id))
id = self.get_node_text(entry_details['pubchem_compound_id'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='pubchem.compound',
id=id))
id = self.get_node_text(entry_details['wikipidia'])
if id:
compound.cross_references.append(
self.get_or_create_object(Resource,
namespace='wikipedia.en',
id=id))
# add to session
db_session.add(compound)
if self.commit_intermediate_results and (i_entry % 100 == 99):
db_session.commit()
if self.verbose:
print(' done')
# commit changes to database
if self.verbose:
print('Saving database ...')
db_session.commit()
if self.verbose:
print(' done')
def test_remove_layer(self):
a = molecule_util.InchiMolecule('InChI=1S/BrH/h1H/p-1')
a.remove_layer('protons')
self.assertEqual(str(a), 'InChI=1S/BrH/h1H')