def get_atom_vector(self):
"""
Returns a NumPy row array describing the number of atoms
from each element (the column index is the atomic number
of that element).
The first column (index=0) contains the number of electrons.
"""
atom_bag = self.get_atom_bag()
if not atom_bag:
return None
atom_vector = np.zeros((Molecule.GetNumberOfElements() + 1),
dtype='int')
for elem, count in atom_bag.iteritems():
if elem in ['R', 'X']:
return None # wildcard compound!
an = Molecule.GetAtomicNum(elem)
if not an:
logging.warning("Unsupported element in (C%05d): %s",
(self.cid, elem))
return None
atom_vector[an] = count
atom_vector[0] = self.get_num_electrons()
return atom_vector
def ToTableString(self):
"""Returns the decomposition as a tabular string."""
spacer = '-' * 50 + '\n'
l = [
'%30s | %2s | %2s | %3s | %s\n' %
("group name", "nH", "z", "nMg", "nodes"), spacer
]
for group, node_sets in self.groups:
if group.hydrogens is None and group.charge is None and group.nMg is None:
for n_set in node_sets:
s = '%30s | | | | %s\n' % \
(group.name, ','.join([str(i) for i in n_set]))
l.append(s)
else:
for n_set in node_sets:
s = '%30s | %2d | %2d | %2d | %s\n' % \
(group.name, group.hydrogens or 0, group.charge or 0, group.nMg or 0,
','.join([str(i) for i in n_set]))
l.append(s)
if self.unassigned_nodes:
l.append('\nUnassigned nodes: \n')
l.append('%10s | %3s | %2s | %10s | %10s\n' %
('index', 'an', 'el', 'valence', 'charge'))
l.append(spacer)
all_atoms = self.mol.GetAtoms()
for i in self.unassigned_nodes:
a = all_atoms[i]
l.append('%10d | %3d | %2s | %10d | %10d\n' %
(i, a.atomicnum, Molecule.GetSymbol(
a.atomicnum), a.heavyvalence, a.formalcharge))
return ''.join(l)
def add_thermodynamics(cursor):
from groups import GroupMissingTrainDataError, GroupDecompositionError
gc = GroupContribution(sqlite_name="gibbs.sqlite", html_name="pathologic")
gc.init()
cursor.execute("DROP TABLE IF EXISTS yeast_inchi2thermo")
cursor.execute(
"CREATE TABLE yeast_inchi2thermo (inchi TEXT, charge INT, nH INT, dG0_f REAL)"
)
cursor.execute("DROP INDEX IF EXISTS yeast_inchi2thermo_idx")
cursor.execute(
"CREATE INDEX yeast_inchi2thermo_idx ON yeast_inchi2thermo (inchi);")
inchi_list = []
for row in cursor.execute("SELECT distinct(inchi) " \
"FROM yeast_species2inchi WHERE inchi IS NOT NULL"):
inchi = row[0]
inchi_list.append(str(inchi))
for inchi in inchi_list:
try:
mol = Molecule.FromInChI(str(inchi))
pmap = gc.Mol2PseudoisomerMap(mol)
for ((z, nH), dG0) in pmap.iteritems():
cursor.execute(
"INSERT INTO yeast_inchi2thermo VALUES(?,?,?,?)",
[inchi, z, nH, dG0])
except (IOError, GroupMissingTrainDataError, GroupDecompositionError):
sys.stderr.write(
"Cannot convert the following InChI to a pybel Molecule")
def run(self):
from toolbox.molecule import Molecule
self.semaphore.acquire()
start_time = time.time()
logging.debug("SMILES: " + self.smiles)
diss_table = Molecule._GetDissociationTable(self.smiles,
fmt='smiles',
mid_pH=default_pH,
min_pKa=0,
max_pKa=14,
T=default_T)
logging.debug("Min charge: %d" % diss_table.min_charge)
logging.debug("Min nH: %d" % diss_table.min_nH)
elapsed_time = time.time() - start_time
self.db_lock.acquire()
db = SqliteDatabase(self.options.db_file)
kegg = Kegg.getInstance()
name = kegg.cid2name(self.cid)
if diss_table is not None:
for row in diss_table.ToDatabaseRow():
db.Insert(self.options.table_name, [self.cid, name] + row)
else:
db.Insert(self.options.table_name, [self.cid, name] + [None] * 10)
del db
self.db_lock.release()
logging.info("Completed C%05d, elapsed time = %.1f sec" %
(self.cid, elapsed_time))
self.semaphore.release()
def run(self):
from toolbox.molecule import Molecule
self.semaphore.acquire()
start_time = time.time()
logging.debug("SMILES: " + self.smiles)
diss_table = Molecule._GetDissociationTable(self.smiles, fmt='smiles',
mid_pH=default_pH, min_pKa=0, max_pKa=14, T=default_T)
logging.debug("Min charge: %d" % diss_table.min_charge)
logging.debug("Min nH: %d" % diss_table.min_nH)
elapsed_time = time.time() - start_time
self.db_lock.acquire()
db = SqliteDatabase(self.options.db_file)
kegg = Kegg.getInstance()
name = kegg.cid2name(self.cid)
if diss_table is not None:
for row in diss_table.ToDatabaseRow():
db.Insert(self.options.table_name, [self.cid, name] + row)
else:
db.Insert(self.options.table_name, [self.cid, name] + [None] * 10)
del db
self.db_lock.release()
logging.info("Completed C%05d, elapsed time = %.1f sec" %
(self.cid, elapsed_time))
self.semaphore.release()
def FromGroupsFile(fp, transformed=False):
"""Factory that initializes a GroupData from a CSV file."""
list_of_groups = []
logging.info('Reading the list of groups from %s' % fp.name)
gid = 0
for row in csv.DictReader(fp):
if row.get('SKIP', False):
logging.warning('Skipping group %s', row.get('NAME'))
continue
group_name = row['NAME']
protons = int(row['PROTONS'])
charge = int(row['CHARGE'])
mgs = int(row['MAGNESIUMS'])
smarts = row['SMARTS']
focal_atoms = FocalSet(row['FOCAL_ATOMS'])
_remark = row['REMARK']
# Check that the smarts are good.
if not Molecule.VerifySmarts(smarts):
raise GroupsDataError('Cannot parse SMARTS: %s' % smarts)
group = Group(gid, group_name, protons, charge, mgs, str(smarts),
focal_atoms)
list_of_groups.append(group)
gid += 1
logging.info('Done reading groups data.')
return GroupsData(list_of_groups, transformed)
def GetMolInput(dissociation):
mols = [
] # a list of pairs of Molecule objects and stoichiometric coefficients
while mols == []:
print 'KEGG ID or SMILES (or Enter to quit):',
s_input = raw_input()
if not s_input:
return []
elif re.findall('C\d\d\d\d\d', s_input) != []:
try:
cid = int(s_input[1:])
mols = [(GetMostAbundantMol(cid, dissociation), 1)]
print "Compound:", mols[0][0].ToInChI()
except ValueError:
print 'syntax error: KEGG compound ID is bad (%s), please try again' % s_input
elif re.findall('R\d\d\d\d\d', s_input) != []:
try:
rid = int(s_input[1:])
reaction = Kegg.getInstance().rid2reaction(rid)
print "Reaction:", str(reaction)
for cid, coeff in reaction.iteritems():
mols += [(GetMostAbundantMol(cid, dissociation), coeff)]
except ValueError:
print 'syntax error: KEGG reaction ID is bad (%s), please try again' % s_input
else:
try:
mols = [(Molecule.FromSmiles(s_input), 1)]
print "Compound:", mols[0][0].ToInChI()
except Exception:
print 'unable to parse SMILES string, please try again'
return mols
def CreateEmptyGroupDecomposition(self):
emptymol = Molecule.FromSmiles("")
decomposition = self.Decompose(emptymol,
ignore_protonations=True,
strict=False)
for i, (group, _node_sets) in enumerate(decomposition.groups):
decomposition.groups[i] = (group, [])
return decomposition
def SetInChI(self, inchi):
if inchi == None:
self.inchi = None
self.mol = None
self.formula = None
self.mass = None
else:
self.inchi = inchi
self.mol = Molecule.FromInChI(inchi)
self.formula = self.mol.GetFormula()
self.mass = self.mol.GetExactMass()
def GetMol(self, nH=None, nMg=0):
from toolbox.molecule import Molecule
if nH is None:
nH = self.min_nH
if (nH, nMg) not in self.mol_dict:
return None
s, mol = self.mol_dict[nH, nMg]
if mol is None:
mol = Molecule.FromSmiles(s)
self.mol_dict[nH, nMg] = (s, mol)
return mol
def EstimateInChI(self, inchi):
mol = Molecule.FromInChI(inchi)
#mol.RemoveHydrogens()
decomposition = self.group_decomposer.Decompose(
mol, ignore_protonations=False, strict=True)
nH = decomposition.Hydrogens()
charge = decomposition.NetCharge()
nMg = decomposition.Magnesiums()
groupvec = decomposition.AsVector()
dG0, ker = self.EstimateGroupVector(groupvec)
return dG0, nH, charge, nMg, ker
def get_nH_and_charge(self):
if not self.mol and self.inchi:
self.mol = Molecule.FromInChI(self.inchi)
if self.mol:
return self.mol.GetHydrogensAndCharge()
# if there is no InChI assume that self.formula is correct and that
# it represents the number of H for the neutral species
atom_bag = self.get_atom_bag()
if not atom_bag:
return None
return atom_bag.get('H', 0), 0
def get_num_electrons(self):
"""Return the putative number of electrons in the molecule."""
mol = self.GetMolecule()
if mol:
return mol.GetNumElectrons()
# if there is no InChI assume that self.formula is correct and that
# the charge is 0.
atom_bag = self.get_atom_bag()
if not atom_bag:
return None
n_protons = 0
for elem, count in atom_bag.iteritems():
n_protons += count * Molecule.GetAtomicNum(elem)
return n_protons
def GetMolecule(self):
"""Gets a Molecule for this compound if possible.
Returns None if no molecular data is available.
"""
if self.mol:
return self.mol
if self.inchi:
self.mol = Molecule.FromInChI(self.inchi)
self.mol.SetTitle(self.name)
return self.mol
raise kegg_errors.KeggParseException(
"C%05d (%s) doesn't have an explicit molecular structure" %
(self.cid, self.name))
def test_dissociation_table(diss,
group_decomposer,
id,
ignore_missing_smiles=False):
if diss is None:
logging.warning('%s: does not appear in the dissociation table' % id)
return
nH, nMg = diss.GetMostAbundantPseudoisomer(pH=default_pH,
I=default_I,
pMg=14,
T=default_T)
if nMg != 0:
logging.warning('%s: default species has nMg = %d' % (id, nMg))
return
smiles = diss.GetSmiles(nH=nH, nMg=0)
if not smiles:
if not ignore_missing_smiles:
logging.warning(
'%s: no SMILES in the dissociation table for nH = %d' %
(id, nH))
return
logging.debug('%s: nH = %d, smiles = %s' % (id, nH, smiles))
mol = Molecule.FromSmiles(smiles)
try:
decomposition = group_decomposer.Decompose(mol,
ignore_protonations=False,
strict=True)
except GroupDecompositionError:
return
groupvec = decomposition.AsVector()
logging.debug("%s: decomposition = %s" % (id, groupvec))
gc_nH = decomposition.Hydrogens()
if nH != gc_nH:
logging.warning(
'%s: nH doesn\'t match: explicit = %d, decomposition = %d' %
(id, nH, gc_nH))
def __init__(self, uid=None, name=None, all_names=None, mass=None,
formula=None, inchi=None, pubchem_id=None, cas=None,
regulates=None, types=None, smiles=None):
self.uid = uid; # UNIQUE-ID
self.name = name # COMMON-NAME
if (self.name):
self.name = re.sub('<.+?>', '', name) # Removing HTML tags
self.all_names = [] # SYNONYMS
if (all_names and len(all_names) > 0):
for s in all_names:
self.all_names.append(re.sub('<.+?>', '', s))
self.mass = mass # MOLECULAR-WEIGHT
self.formula = formula # CHEMICAL-FORMULA
self.inchi = inchi if inchi != None else "" # INCHI
self.pubchem_id = None # Parsed from DBLINKS
self.cas = "" # Parsed from DBLINKS
self.regulates = regulates if regulates != None else [] # REGULATES
self.types = types if types != None else [] # TYPES
self.smiles = smiles if smiles != None else "" # SMILES
if (smiles and not inchi):
self.inchi = Molecule.Smiles2InChI(smiles)
def ConvertFormation2Reaction(self, output_fname):
logging.info("Converting all formation energies to reactions")
output_csv = csv.writer(open(output_fname, 'w'))
# keep the format used for TECRDB
output_csv.writerow(
('ref', 'ID', 'method', 'eval', 'EC', 'name', 'kegg_reaction',
'reaction', 'dG0\'', 'T', 'I', 'pH', 'pMg'))
atom2cid = {}
for atom, (name, stoich) in KeggObservation.ATOM2ELEMENT.iteritems():
cid, _, _ = self.kegg.name2cid(name, 0)
if cid is None:
raise Exception(
"Cannot find the element %s in the KEGG database" % name)
atom2cid[atom] = (cid, stoich)
#output_csv.writerow(('element',
# 'C%05d' % cid, 'formation', 'A', '',
# 'formation of %s' % self.kegg.cid2name(cid),
# "C%05d" % cid,
# name, 0, self.T, self.I, self.pH, self.pMg))
for label in ['training', 'testing']:
ptable = PsuedoisomerTableThermodynamics.FromCsvFile(
self.FormationEnergyFileName, label=label)
for cid in ptable.get_all_cids():
pmatrix = ptable.cid2PseudoisomerMap(cid).ToMatrix()
if len(pmatrix) != 1:
raise Exception("multiple training species for C%05d" %
cid)
nH, _charge, nMg, dG0 = pmatrix[0]
diss_table = dissociation.GetDissociationTable(cid, False)
if diss_table is None:
continue
diss_table.SetFormationEnergyByNumHydrogens(dG0, nH, nMg)
dG0_prime = diss_table.Transform(pH=self.pH,
I=self.I,
pMg=self.pMg,
T=self.T)
ref = ptable.cid2SourceString(cid)
atom_bag = self.kegg.cid2atom_bag(cid)
if not atom_bag:
continue
ne = self.kegg.cid2num_electrons(cid)
elem_ne = 0
sparse = {cid: 1}
for elem, count in atom_bag.iteritems():
if elem == 'H':
continue
elem_ne += count * Molecule.GetAtomicNum(elem)
elem_cid, elem_coeff = atom2cid[elem]
sparse.setdefault(elem_cid, 0)
sparse[elem_cid] += -count * elem_coeff
# use the H element to balance the electrons in the formation
# reactions (we don't need to balance protons since this is
# a biochemical reaction, so H+ are 'free').
H_cid, H_coeff = atom2cid['H']
sparse[H_cid] = (elem_ne - ne) * H_coeff
reaction = Reaction(
"formation of %s" % self.kegg.cid2name(cid), sparse)
output_csv.writerow(
(ref, 'C%05d' % cid, 'formation', 'A', '',
'formation of %s' % self.kegg.cid2name(cid),
reaction.FullReactionString(),
reaction.FullReactionString(show_cids=False),
'%.2f' % dG0_prime, self.T, self.I, self.pH, self.pMg))
def CalculateThermo():
parser = MakeOpts()
options, _ = parser.parse_args(sys.argv)
pH, I, pMg, T = options.pH, options.I, options.pMg, options.T
db = SqliteDatabase('../res/gibbs.sqlite')
G = GroupContribution(db=db)
G.init()
ignore_protonations = False
list_of_mols = []
if options.smiles:
list_of_mols.append({
'id': options.smiles,
'mol': options.smiles,
'format': 'smiles'
})
elif options.inchi:
list_of_mols.append({
'id': options.inchi,
'mol': options.inchi,
'format': 'inchi'
})
elif options.csv_input_filename:
for row in csv.DictReader(open(options.csv_input_filename, 'r')):
if "InChI" in row:
list_of_mols.append({
'id': row["ID"],
'mol': row["InChI"],
'format': 'inchi'
})
elif "smiles" in row:
list_of_mols.append({
'id': row["ID"],
'mol': row["smiles"],
'format': 'smiles'
})
else:
raise Exception(
"There must be one molecular ID column: InChI or smiles")
else:
parser.error("must use either -s or -c option")
if options.biochemical:
print(
"Calculating biochemical formation energies for %s compounds"
" at pH = %.1f, I = %.2f, pMg = %.1f, T = %.2f" %
(len(list_of_mols), pH, I, pMg, T))
else:
print("Calculating chemical formation energies for %s compounds" %
len(list_of_mols))
rowdicts = []
for mol_dict in list_of_mols:
mol_id = mol_dict['id']
diss_table = Molecule._GetDissociationTable(mol_dict['mol'],
fmt=mol_dict['format'])
try:
mol = diss_table.GetMostAbundantMol(pH, I, pMg, T) or \
diss_table.GetAnyMol()
if mol is None:
raise Exception("Cannot convert input string to Molecule: " +
mol_dict['mol'])
decomposition = G.Mol2Decomposition(
mol, ignore_protonations=ignore_protonations)
groupvec = decomposition.AsVector()
dG0 = G.groupvec2val(groupvec)
nH = decomposition.Hydrogens()
nMg = decomposition.Magnesiums()
diss_table.SetFormationEnergyByNumHydrogens(dG0, nH, nMg)
pmap = diss_table.GetPseudoisomerMap()
if options.biochemical:
dG0_prime = pmap.Transform(pH, pMg, I, T)
rowdicts.append({
'ID': mol_id,
'pH': pH,
'I': I,
'pMg': pMg,
'dG0\'': "%.1f" % dG0_prime,
'groupvec': str(groupvec)
})
else:
for p_nH, p_z, p_nMg, p_dG0 in pmap.ToMatrix():
rowdicts.append({
'ID': mol_id,
'nH': p_nH,
'charge': p_z,
'nMg': p_nMg,
'dG0': "%.1f" % p_dG0,
'groupvec': str(groupvec)
})
except GroupDecompositionError:
rowdicts.append({'ID': mol_id, 'error': "cannot decompose"})
except GroupMissingTrainDataError:
rowdicts.append({
'ID': mol_id,
'groupvec': str(groupvec),
'error': "missing training data"
})
if options.csv_output_filename is not None:
out_fp = open(options.csv_output_filename, 'w')
print "writing results to %s ... " % options.csv_output_filename
else:
out_fp = sys.stdout
if options.biochemical:
titles = ['ID', 'error', 'pH', 'I', 'pMg', 'dG0\'', 'groupvec']
else:
titles = ['ID', 'error', 'nH', 'nMg', 'charge', 'dG0', 'groupvec']
csv_writer = csv.DictWriter(out_fp, titles)
csv_writer.writeheader()
csv_writer.writerows(rowdicts)
import unittest
from pygibbs.groups_data import GroupsData
from pygibbs import group_decomposition
from toolbox.molecule import Molecule
import logging
PHOSPHATE = Molecule.FromSmiles('[O-]P([O-])(=O)O')
ATP = Molecule.FromSmiles(
'C1=NC2=C(C(=N1)N)N=CN2C3C(C(C(O3)COP(=O)([O-])OP(=O)([O-])OP(=O)([O-])O)O)O'
)
A4P = Molecule.FromSmiles(
'C1=NC2=C(C(=N1)N)N=CN2C3C(C(C(O3)COP(=O)([O-])OP(=O)([O-])OP(=O)([O-])OP(=O)([O-])O)O)O'
)
class GroupsDecompositionTest(unittest.TestCase):
"""Tests for GroupsDecomposition"""
def setUp(self):
self.groups_decomposer = group_decomposition.GroupDecomposer.FromGroupsFile(
open('../data/thermodynamics/groups_species.csv', 'r'))
def testFindPhosphateChains(self):
ps = group_decomposition.GroupDecomposer.FindPhosphateChains(
PHOSPHATE, ignore_protonations=False)
for unused_grp, l in ps:
self.assertTrue(not l)
mk_ps_dict = lambda ps: dict((key, l) for key, l in ps)
mk_ps_string = lambda ps: ', '.join(
["%s x %d" % (str(key), len(l)) for key, l in ps if l != []])
logging.error(e)
continue
return
atp = 'C1=NC2=C(C(=N1)N)N=CN2C3C(C(C(O3)COP(=O)(O)OP(=O)(O)OP(=O)(O)O)O)O'
coa = 'C1C=CN(C=C1C(=O)N)C2C(C(C(O2)COP(=O)(O)OP(=O)(O)OCC3C(C(C(O3)N4C=NC5=C4N=CN=C5N)O)O)O)O'
glucose = 'C(C1C(C(C(C(O1)O)O)O)O)O'
mgatp = 'C([C@@H]1[C@H]([C@H]([C@H](n2cnc3c(N)[nH+]cnc23)O1)O)O)OP(=O)([O-])OP(=O)([O-])OP(=O)([O-])[O-].[Mg+2].[Mg+2]'
#smiless = [
# ('ATP', atp),
# ('CoA', coa), ('Glucose', glucose), ('MgAtp', mgatp),
# ]
smiless = [('ATP', atp)]
mols = [(name, Molecule.FromSmiles(s)) for name, s in smiless]
for name, mol in mols:
print name
decomposition = decomposer.Decompose(mol)
print decomposition.ToTableString()
print 'Group count', decomposition.group_count
print 'Net charge', decomposition.net_charge
print 'Hydrogens', decomposition.hydrogens
print 'Magnesiums', decomposition.magnesiums
print 'Group Vector:'
print decomposition.AsVector()
print 'Pseudoisomer Vectors:'
for v in decomposition.PseudoisomerVectors():
def main():
mol = Molecule.FromSmiles('C(O)(=O)C(=O)O')
emp = EnzymeMarketplace()
print emp.React(mol)
sum_conc = 0
for Ka_subset in itertools.combinations(Ka_list, i+1): # all choices of i values from the Ka list
sum_conc += np.prod(Ka_subset)
relative_conc.append(sum_conc)
Ka_i = relative_conc[i+1] / relative_conc[i]
transformed_pKas.append(-np.log10(Ka_i))
return transformed_pKas
if __name__ == "__main__":
diss_table_example = [4.0, 4.0, 4.0]
new_diss_table = _TransformMultiples(diss_table_example)
print diss_table_example
print new_diss_table
from toolbox.molecule import Molecule
compound_list = [('glycine', 'C(=O)(O)CN'),
('CO2', 'O=C=O'),
('ATP', 'Nc1ncnc2n(cnc12)C1OC(COP([O-])(=O)OP([O-])(=O)OP(O)([O-])=O)C(O)C1O'),
('3-Ketoarabinitol', 'OCC(O)C(C(O)CO)=O')]
for name, smiles in compound_list:
diss_table1, major_ms = GetDissociationConstants(smiles, transform_multiples=False)
diss_table2, major_ms = GetDissociationConstants(smiles, transform_multiples=True)
m = Molecule.FromSmiles(major_ms)
print name, m.ToInChI()
for i in xrange(len(diss_table1)):
print "%.2f %.2f" % (diss_table1[i][0], diss_table2[i][0])
def CalculateThermo():
options, _ = MakeOpts().parse_args(sys.argv)
if options.csv_output_filename is not None:
out_fp = open(options.csv_output_filename, 'w')
print "writing results to %s ... " % options.csv_output_filename
else:
out_fp = sys.stdout
csv_writer = csv.writer(out_fp)
csv_writer.writerow(
['ID', 'error', 'nH', 'nMg', 'charge', 'dG0', 'kernel'])
db = SqliteDatabase('../res/gibbs.sqlite', 'w')
ugc = UnifiedGroupContribution(db)
ugc.LoadGroups(True)
ugc.LoadObservations(True)
ugc.LoadGroupVectors(True)
ugc.LoadData(True)
result_dict = ugc._GetContributionData(ugc.S.copy(), ugc.cids,
ugc.b.copy(), ugc.anchored)
g_pgc = result_dict['group_contributions']
P_L_pgc = result_dict['pgc_conservations']
sdfile = pybel.readfile("sdf", options.sdf_input_filename)
for m in sdfile:
try:
try:
mol = Molecule.FromOBMol(m.OBMol)
except OpenBabelError:
raise UnknownReactionEnergyError(
"Cannot convert to OBMol object")
mol.title = m.title
mol.RemoveHydrogens()
if mol.GetNumAtoms() > 200:
raise UnknownReactionEnergyError(
"Compound contains more than 200 atoms (n = %d)" %
mol.GetNumAtoms())
try:
decomposition = ugc.group_decomposer.Decompose(
mol, ignore_protonations=False, strict=True)
except GroupDecompositionError:
raise UnknownReactionEnergyError("cannot decompose")
groupvec = decomposition.AsVector()
gv = np.matrix(groupvec.Flatten())
dG0 = float(g_pgc * gv.T)
nH = decomposition.Hydrogens()
nMg = decomposition.Magnesiums()
ker = list((P_L_pgc * gv.T).round(10).flat)
try:
diss_table = mol.GetDissociationTable()
diss_table.SetFormationEnergyByNumHydrogens(dG0=dG0,
nH=nH,
nMg=nMg)
except MissingDissociationConstantError:
raise UnknownReactionEnergyError("missing pKa data")
pmap = diss_table.GetPseudoisomerMap()
for p_nH, p_z, p_nMg, p_dG0 in pmap.ToMatrix():
csv_writer.writerow([
m.title, None, p_nH, p_z, p_nMg,
round(p_dG0, 1),
str(ker)
])
except UnknownReactionEnergyError as e:
csv_writer.writerow(
[m.title, str(e), None, None, None, None, None])
out_fp.flush()
def CalculateThermo():
parser = MakeOpts()
options, _ = parser.parse_args(sys.argv)
pH, I, pMg, T = options.pH, options.I, options.pMg, options.T
db = SqliteDatabase('../res/gibbs.sqlite')
G = GroupContribution(db=db)
G.init()
ignore_protonations = False
list_of_mols = []
if options.smiles:
list_of_mols.append({'id':options.smiles, 'mol':options.smiles,
'format':'smiles'})
elif options.inchi:
list_of_mols.append({'id':options.inchi, 'mol':options.inchi,
'format':'inchi'})
elif options.csv_input_filename:
for row in csv.DictReader(open(options.csv_input_filename, 'r')):
if "InChI" in row:
list_of_mols.append({'id':row["ID"], 'mol':row["InChI"],
'format':'inchi'})
elif "smiles" in row:
list_of_mols.append({'id':row["ID"], 'mol':row["smiles"],
'format':'smiles'})
else:
raise Exception("There must be one molecular ID column: InChI or smiles")
else:
parser.error("must use either -s or -c option")
if options.biochemical:
print ("Calculating biochemical formation energies for %s compounds"
" at pH = %.1f, I = %.2f, pMg = %.1f, T = %.2f" %
(len(list_of_mols), pH, I, pMg, T))
else:
print ("Calculating chemical formation energies for %s compounds" %
len(list_of_mols))
rowdicts = []
for mol_dict in list_of_mols:
mol_id = mol_dict['id']
diss_table = Molecule._GetDissociationTable(mol_dict['mol'],
fmt=mol_dict['format'])
try:
mol = diss_table.GetMostAbundantMol(pH, I, pMg, T) or \
diss_table.GetAnyMol()
if mol is None:
raise Exception("Cannot convert input string to Molecule: " +
mol_dict['mol'])
decomposition = G.Mol2Decomposition(mol,
ignore_protonations=ignore_protonations)
groupvec = decomposition.AsVector()
dG0 = G.groupvec2val(groupvec)
nH = decomposition.Hydrogens()
nMg = decomposition.Magnesiums()
diss_table.SetFormationEnergyByNumHydrogens(dG0, nH, nMg)
pmap = diss_table.GetPseudoisomerMap()
if options.biochemical:
dG0_prime = pmap.Transform(pH, pMg, I, T)
rowdicts.append({'ID':mol_id, 'pH':pH, 'I':I, 'pMg':pMg,
'dG0\'':"%.1f" % dG0_prime, 'groupvec':str(groupvec)})
else:
for p_nH, p_z, p_nMg, p_dG0 in pmap.ToMatrix():
rowdicts.append({'ID':mol_id, 'nH':p_nH, 'charge':p_z, 'nMg':p_nMg,
'dG0':"%.1f" % p_dG0, 'groupvec':str(groupvec)})
except GroupDecompositionError:
rowdicts.append({'ID':mol_id, 'error':"cannot decompose"})
except GroupMissingTrainDataError:
rowdicts.append({'ID':mol_id, 'groupvec':str(groupvec),
'error':"missing training data"})
if options.csv_output_filename is not None:
out_fp = open(options.csv_output_filename, 'w')
print "writing results to %s ... " % options.csv_output_filename
else:
out_fp = sys.stdout
if options.biochemical:
titles = ['ID', 'error', 'pH', 'I', 'pMg', 'dG0\'', 'groupvec']
else:
titles = ['ID', 'error', 'nH', 'nMg', 'charge', 'dG0', 'groupvec']
csv_writer = csv.DictWriter(out_fp, titles)
csv_writer.writeheader()
csv_writer.writerows(rowdicts)