def generate_isotopomers(spc, N=1):
"""
Generate all isotopomers of the parameter species by adding max. N carbon isotopes to the
atoms of the species.
"""
mol = spc.molecule[0]
isotope = getElement(6, 13)
mols = []
add_isotope(0, N, mol, mols, isotope)
spcs = []
for isomol in mols:
isotopomer = Species(molecule=[isomol], thermo=deepcopy(spc.thermo), transportData=spc.transportData, reactive=spc.reactive)
isotopomer.generate_resonance_structures(keep_isomorphic=True)
spcs.append(isotopomer)
# do not retain identical species:
filtered = []
while spcs:
candidate = spcs.pop()
unique = True
for isotopomer in filtered:
if isotopomer.isIsomorphic(candidate):
unique = False
break
if unique: filtered.append(candidate)
if spc.thermo:
for isotopomer in filtered:
correct_entropy(isotopomer, spc)
return filtered
def testThermoForMonocyclicAndPolycyclicSameMolecule(self):
"""
Test a molecule that has both a polycyclic and a monocyclic ring in the same molecule
"""
spec = Species().fromSMILES('C(CCC1C2CCC1CC2)CC1CCC1')
spec.generateResonanceIsomers()
thermo = self.database.getThermoDataFromGroups(spec)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(
thermo)
self.assertEqual(len(ringGroups), 1)
self.assertEqual(len(polycyclicGroups), 1)
expected_matchedRingsLabels = ['Cyclobutane']
expected_matchedRings = [
self.database.groups['ring'].entries[label]
for label in expected_matchedRingsLabels
]
self.assertEqual(set(ringGroups), set(expected_matchedRings))
expected_matchedPolyringsLabels = ['s3_5_5_ane']
expected_matchedPolyrings = [
self.database.groups['polycyclic'].entries[label]
for label in expected_matchedPolyringsLabels
]
self.assertEqual(set(polycyclicGroups), set(expected_matchedPolyrings))
def generate_isotopomers(spc, N=1):
"""
Generate all isotopomers of the parameter species by adding max. N carbon isotopes to the
atoms of the species.
"""
mol = spc.molecule[0]
isotope = getElement(6, 13)
mols = []
add_isotope(0, N, mol, mols, isotope)
spcs = []
for isomol in mols:
isotopomer = Species(molecule=[isomol], thermo=deepcopy(spc.thermo), transportData=spc.transportData, reactive=spc.reactive)
isotopomer.generate_resonance_structures(keep_isomorphic=True)
spcs.append(isotopomer)
# do not retain identical species:
filtered = []
while spcs:
candidate = spcs.pop()
unique = True
for isotopomer in filtered:
if isotopomer.isIsomorphic(candidate):
unique = False
break
if unique: filtered.append(candidate)
if spc.thermo:
for isotopomer in filtered:
correct_entropy(isotopomer, spc)
return filtered
def loadSpeciesDictionary(path):
"""
Very similar method with `loadSpeciesDictionary` in RMG-Py. Here this method
doesn't generateResonanceIsomers for the created species, will just keep it
as it is.
"""
speciesDict = {}
inerts = [Species().fromSMILES(inert) for inert in ('[He]', '[Ne]', 'N#N', '[Ar]')]
with open(path, 'r') as f:
adjlist = ''
for line in f:
if line.strip() == '' and adjlist.strip() != '':
# Finish this adjacency list
species = Species().fromAdjacencyList(adjlist)
label = species.label
for inert in inerts:
if inert.isIsomorphic(species):
species.reactive = False
break
speciesDict[label] = species
adjlist = ''
else:
if "InChI" in line:
line = line.split()[0] + '\n'
if '//' in line:
index = line.index('//')
line = line[0:index]
adjlist += line
return speciesDict
def testGetRingGroupsFromComments(self):
"""
Test that getRingGroupsFromComments method works for fused polycyclics.
"""
from rmgpy.thermo.thermoengine import generateThermoData
# set-up RMG object
rmg = RMG()
# load kinetic database and forbidden structures
rmg.database = RMGDatabase()
path = os.path.join(settings['database.directory'])
# forbidden structure loading
rmg.database.loadThermo(os.path.join(path, 'thermo'))
smi = 'C12C(C3CCC2C3)C4CCC1C4'#two norbornane rings fused together
spc = Species().fromSMILES(smi)
spc.thermo = generateThermoData(spc)
thermodb = rmg.database.thermo
thermodb.getRingGroupsFromComments(spc.thermo)
import rmgpy.data.rmg
rmgpy.data.rmg.database = None
def testPolycyclicPicksBestThermo(self):
"""
Test that RMG prioritizes thermo correctly and chooses the thermo from the isomer which
has a non generic polycyclic ring correction
"""
spec = Species().fromSMILES('C1=C[C]2CCC=C2C1')
spec.generateResonanceIsomers()
thermoDataList = []
for molecule in spec.molecule:
thermo = self.database.estimateRadicalThermoViaHBI(molecule, self.database.computeGroupAdditivityThermo)
thermoDataList.append(thermo)
thermoDataList.sort(key=lambda x: x.getEnthalpy(298))
most_stable_thermo = thermoDataList[0]
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(most_stable_thermo)
selected_thermo = self.database.getThermoDataFromGroups(spec)
self.assertNotEqual(selected_thermo, thermoDataList)
selected_ringGroups, selected_polycyclicGroups = self.database.getRingGroupsFromComments(selected_thermo)
# The group used to estimate the most stable thermo is the generic polycyclic group and
# therefore is not selected. Note that this unit test will have to change if the correction is fixed later.
self.assertEqual(polycyclicGroups[0].label, 'PolycyclicRing')
self.assertEqual(selected_polycyclicGroups[0].label, 'C12CCC=C1CC=C2')
def testPolycyclicPicksBestThermo(self):
"""
Test that RMG prioritizes thermo correctly and chooses the thermo from the isomer which
has a non generic polycyclic ring correction
"""
spec = Species().fromSMILES('C1=C[C]2CCC=C2C1')
spec.generateResonanceIsomers()
thermoDataList = []
for molecule in spec.molecule:
thermo = self.database.estimateRadicalThermoViaHBI(
molecule, self.database.computeGroupAdditivityThermo)
thermoDataList.append(thermo)
thermoDataList.sort(key=lambda x: x.getEnthalpy(298))
most_stable_thermo = thermoDataList[0]
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(
most_stable_thermo)
selected_thermo = self.database.getThermoDataFromGroups(spec)
self.assertNotEqual(selected_thermo, thermoDataList)
selected_ringGroups, selected_polycyclicGroups = self.database.getRingGroupsFromComments(
selected_thermo)
# The group used to estimate the most stable thermo is the generic polycyclic group and
# therefore is not selected. Note that this unit test will have to change if the correction is fixed later.
self.assertEqual(polycyclicGroups[0].label, 'PolycyclicRing')
self.assertEqual(selected_polycyclicGroups[0].label, 'C12CCC=C1CC=C2')
def testNewThermoGeneration(self):
"""
Test that the new ThermoDatabase generates appropriate thermo data.
"""
for smiles, symm, H298, S298, Cp300, Cp400, Cp500, Cp600, Cp800, Cp1000, Cp1500 in self.testCases:
Cplist = [Cp300, Cp400, Cp500, Cp600, Cp800, Cp1000, Cp1500]
molecule=Molecule(SMILES=smiles)
species = Species(molecule=molecule)
species.generateResonanceIsomers()
species.molecule[0]
thermoData = self.database.getThermoDataFromGroups(species)
molecule = species.molecule[0]
for mol in species.molecule[1:]:
thermoData0 = self.database.getAllThermoData(Species(molecule=[mol]))[0][0]
for data in self.database.getAllThermoData(Species(molecule=[mol]))[1:]:
if data.getEnthalpy(298) < thermoData0.getEnthalpy(298):
thermoData0 = data
if thermoData0.getEnthalpy(298) < thermoData.getEnthalpy(298):
thermoData = thermoData0
molecule = mol
self.assertAlmostEqual(H298, thermoData.getEnthalpy(298) / 4184, places=1, msg="H298 error for {0}".format(smiles))
self.assertAlmostEqual(S298, thermoData.getEntropy(298) / 4.184, places=1, msg="S298 error for {0}".format(smiles))
for T, Cp in zip(self.Tlist, Cplist):
self.assertAlmostEqual(Cp, thermoData.getHeatCapacity(T) / 4.184, places=1, msg="Cp{1} error for {0}".format(smiles,T))
def testGetRingGroupsFromComments(self):
"""
Test that getRingGroupsFromComments method works for fused polycyclics.
"""
from rmgpy.thermo.thermoengine import generateThermoData
# set-up RMG object
rmg = RMG()
# load kinetic database and forbidden structures
rmg.database = RMGDatabase()
path = os.path.join(settings['database.directory'])
# forbidden structure loading
rmg.database.loadThermo(os.path.join(path, 'thermo'))
smi = 'C12C(C3CCC2C3)C4CCC1C4' #two norbornane rings fused together
spc = Species().fromSMILES(smi)
spc.thermo = generateThermoData(spc)
thermodb = rmg.database.thermo
thermodb.getRingGroupsFromComments(spc.thermo)
import rmgpy.data.rmg
rmgpy.data.rmg.database = None
def testReact(self):
"""
Test that reaction generation from the available families works.
"""
spcA = Species().fromSMILES('[OH]')
spcs = [Species().fromSMILES('CC'), Species().fromSMILES('[CH3]')]
reactionList = list(react(spcA, spcs))
self.assertIsNotNone(reactionList)
self.assertTrue(all([isinstance(rxn, TemplateReaction) for rxn in reactionList]))
def testAddPolyRingCorrectionThermoDataFromHeuristicUsingPyrene(self):
# create testing molecule: Pyrene with two ring of aromatic version
# the other two ring of kekulized version
#
# creating it seems not natural in RMG, that's because
# RMG cannot parse the adjacencyList of that isomer correctly
# so here we start with pyrene radical and get the two aromatic ring isomer
# then saturate it.
smiles = '[C]1C=C2C=CC=C3C=CC4=CC=CC=1C4=C23'
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
mols = []
for mol in spe.molecule:
sssr0 = mol.getSmallestSetOfSmallestRings()
aromaticRingNum = 0
for sr0 in sssr0:
sr0mol = Molecule(atoms=sr0)
if isAromaticRing(sr0mol):
aromaticRingNum += 1
if aromaticRingNum == 2:
mol.saturate()
mols.append(mol)
ringGroupLabels = []
polycyclicGroupLabels = []
for mol in mols:
polyring = mol.getDisparateRings()[1][0]
thermoData = ThermoData(
Tdata=([300, 400, 500, 600, 800, 1000, 1500], "K"),
Cpdata=([0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], "J/(mol*K)"),
H298=(0.0, "kJ/mol"),
S298=(0.0, "J/(mol*K)"),
)
self.database._ThermoDatabase__addPolyRingCorrectionThermoDataFromHeuristic(
thermoData, polyring)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(
thermoData)
ringGroupLabels += [ringGroup.label for ringGroup in ringGroups]
polycyclicGroupLabels += [
polycyclicGroup.label for polycyclicGroup in polycyclicGroups
]
self.assertIn('Benzene', ringGroupLabels)
self.assertIn('six-inringtwodouble-12', ringGroupLabels)
self.assertIn('Cyclohexene', ringGroupLabels)
self.assertIn('1,3-Cyclohexadiene', ringGroupLabels)
self.assertIn('s2_6_6_ben_ene_1', polycyclicGroupLabels)
self.assertIn('s2_6_6_ben_ene_2', polycyclicGroupLabels)
self.assertIn('s2_6_6_naphthalene', polycyclicGroupLabels)
def testGetRingGroupsFromComments(self):
"""
Test that getRingGroupsFromComments method works for fused polycyclics.
"""
from rmgpy.thermo.thermoengine import generateThermoData
smi = 'C12C(C3CCC2C3)C4CCC1C4'#two norbornane rings fused together
spc = Species().fromSMILES(smi)
spc.thermo = generateThermoData(spc)
self.database.getRingGroupsFromComments(spc.thermo)
def testSpeciesThermoGenerationHBIGAV(self):
"""Test thermo generation for species objects.
Ensure that molecule list is only reordered, and not changed after group additivity"""
spec = Species().fromSMILES('CCC[CH]c1ccccc1')
spec.generateResonanceIsomers()
initial = list(spec.molecule) # Make a copy of the list
thermo = self.database.getThermoData(spec)
self.assertEqual(len(initial), len(spec.molecule))
self.assertEqual(set(initial), set(spec.molecule))
self.assertTrue('group additivity' in thermo.comment, 'Thermo not found from GAV, test purpose not fulfilled.')
def testGetRingGroupsFromComments(self):
"""
Test that getRingGroupsFromComments method works for fused polycyclics.
"""
from rmgpy.thermo.thermoengine import generateThermoData
smi = 'C12C(C3CCC2C3)C4CCC1C4' #two norbornane rings fused together
spc = Species().fromSMILES(smi)
spc.thermo = generateThermoData(spc)
self.database.getRingGroupsFromComments(spc.thermo)
def testBicyclicDecompositionForPolyringUsingPyrene(self):
# create testing molecule: Pyrene with two ring of aromatic version
# the other two ring of kekulized version
#
# creating it seems not natural in RMG, that's because
# RMG cannot parse the adjacencyList of that isomer correctly
# so here we start with pyrene radical and get the two aromatic ring isomer
# then saturate it.
smiles = '[C]1C=C2C=CC=C3C=CC4=CC=CC=1C4=C23'
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
for mol in spe.molecule:
sssr0 = mol.getSmallestSetOfSmallestRings()
aromaticRingNum = 0
for sr0 in sssr0:
sr0mol = Molecule(atoms=sr0)
if isAromaticRing(sr0mol):
aromaticRingNum += 1
if aromaticRingNum == 2:
break
mol.saturate()
# extract polyring from the molecule
polyring = mol.getDisparateRings()[1][0]
bicyclicList, ringOccurancesDict = bicyclicDecompositionForPolyring(
polyring)
# 1st test: number of cores
self.assertEqual(len(bicyclicList), 5)
# 2nd test: ringOccurancesDict
ringInCoreOccurances = sorted(ringOccurancesDict.values())
expectedRingInCoreOccurances = [2, 2, 3, 3]
self.assertEqual(ringInCoreOccurances, expectedRingInCoreOccurances)
# 3rd test: size of each bicyclic core
bicyclicSizes = sorted(
[len(bicyclic.atoms) for bicyclic in bicyclicList])
expectedBicyclicSizes = [10, 10, 10, 10, 10]
self.assertEqual(bicyclicSizes, expectedBicyclicSizes)
# 4th test: bond info for members of each core
aromaticBondNumInBicyclics = []
for bicyclic in bicyclicList:
aromaticBondNum = len(findAromaticBondsFromSubMolecule(bicyclic))
aromaticBondNumInBicyclics.append(aromaticBondNum)
aromaticBondNumInBicyclics = sorted(aromaticBondNumInBicyclics)
expectedAromaticBondNumInBicyclics = [0, 6, 6, 6, 11]
self.assertEqual(aromaticBondNumInBicyclics,
expectedAromaticBondNumInBicyclics)
def testAddPolyRingCorrectionThermoDataFromHeuristicUsingPyrene(self):
# create testing molecule: Pyrene with two ring of aromatic version
# the other two ring of kekulized version
#
# creating it seems not natural in RMG, that's because
# RMG cannot parse the adjacencyList of that isomer correctly
# so here we start with pyrene radical and get the two aromatic ring isomer
# then saturate it.
smiles = '[C]1C=C2C=CC=C3C=CC4=CC=CC=1C4=C23'
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
mols = []
for mol in spe.molecule:
sssr0 = mol.getSmallestSetOfSmallestRings()
aromaticRingNum = 0
for sr0 in sssr0:
sr0mol = Molecule(atoms=sr0)
if isAromaticRing(sr0mol):
aromaticRingNum += 1
if aromaticRingNum == 2:
mol.saturate()
mols.append(mol)
ringGroupLabels = []
polycyclicGroupLabels = []
for mol in mols:
polyring = mol.getDisparateRings()[1][0]
thermoData = ThermoData(
Tdata = ([300,400,500,600,800,1000,1500],"K"),
Cpdata = ([0.0,0.0,0.0,0.0,0.0,0.0,0.0],"J/(mol*K)"),
H298 = (0.0,"kJ/mol"),
S298 = (0.0,"J/(mol*K)"),
)
self.database._ThermoDatabase__addPolyRingCorrectionThermoDataFromHeuristic(
thermoData, polyring)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(thermoData)
ringGroupLabels += [ringGroup.label for ringGroup in ringGroups]
polycyclicGroupLabels += [polycyclicGroup.label for polycyclicGroup in polycyclicGroups]
self.assertIn('Benzene', ringGroupLabels)
self.assertIn('six-inringtwodouble-12', ringGroupLabels)
self.assertIn('Cyclohexene', ringGroupLabels)
self.assertIn('1,3-Cyclohexadiene', ringGroupLabels)
self.assertIn('s2_6_6_ben_ene_1', polycyclicGroupLabels)
self.assertIn('s2_6_6_ben_ene_2', polycyclicGroupLabels)
self.assertIn('s2_6_6_naphthalene', polycyclicGroupLabels)
def testSpeciesThermoGenerationHBIGAV(self):
"""Test thermo generation for species objects.
Ensure that molecule list is only reordered, and not changed after group additivity"""
spec = Species().fromSMILES('CCC[CH]c1ccccc1')
spec.generateResonanceIsomers()
initial = list(spec.molecule) # Make a copy of the list
thermo = self.database.getThermoData(spec)
self.assertEqual(len(initial), len(spec.molecule))
self.assertEqual(set(initial), set(spec.molecule))
self.assertTrue(
'group additivity' in thermo.comment,
'Thermo not found from GAV, test purpose not fulfilled.')
def testBicyclicDecompositionForPolyringUsingPyrene(self):
# create testing molecule: Pyrene with two ring of aromatic version
# the other two ring of kekulized version
#
# creating it seems not natural in RMG, that's because
# RMG cannot parse the adjacencyList of that isomer correctly
# so here we start with pyrene radical and get the two aromatic ring isomer
# then saturate it.
smiles = '[C]1C=C2C=CC=C3C=CC4=CC=CC=1C4=C23'
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
for mol in spe.molecule:
sssr0 = mol.getSmallestSetOfSmallestRings()
aromaticRingNum = 0
for sr0 in sssr0:
sr0mol = Molecule(atoms=sr0)
if isAromaticRing(sr0mol):
aromaticRingNum += 1
if aromaticRingNum == 2:
break
mol.saturate()
# extract polyring from the molecule
polyring = mol.getDisparateRings()[1][0]
bicyclicList, ringOccurancesDict = bicyclicDecompositionForPolyring(polyring)
# 1st test: number of cores
self.assertEqual(len(bicyclicList), 5)
# 2nd test: ringOccurancesDict
ringInCoreOccurances = sorted(ringOccurancesDict.values())
expectedRingInCoreOccurances = [2, 2, 3, 3]
self.assertEqual(ringInCoreOccurances, expectedRingInCoreOccurances)
# 3rd test: size of each bicyclic core
bicyclicSizes = sorted([len(bicyclic.atoms) for bicyclic in bicyclicList])
expectedBicyclicSizes = [10, 10, 10, 10, 10]
self.assertEqual(bicyclicSizes, expectedBicyclicSizes)
# 4th test: bond info for members of each core
aromaticBondNumInBicyclics = []
for bicyclic in bicyclicList:
aromaticBondNum = len(findAromaticBondsFromSubMolecule(bicyclic))
aromaticBondNumInBicyclics.append(aromaticBondNum)
aromaticBondNumInBicyclics = sorted(aromaticBondNumInBicyclics)
expectedAromaticBondNumInBicyclics = [0, 6, 6, 6, 11]
self.assertEqual(aromaticBondNumInBicyclics, expectedAromaticBondNumInBicyclics)
def generate():
"""
Test that reaction generation from the available families works.
"""
load()
spcA = Species().fromSMILES('[OH]')
spcs = [Species().fromSMILES('CC'), Species().fromSMILES('[CH3]')]
reactionList = list(react(spcA, spcs))
if not reactionList: return False
for rxn in reactionList:
if not isinstance(rxn, TemplateReaction): return False
return True
def testAddPolyRingCorrectionThermoDataFromHeuristicUsingAromaticTricyclic(
self):
# create testing molecule
#
# creating it seems not natural in RMG, that's because
# RMG cannot parse the adjacencyList of that isomer correctly
# so here we start with kekulized version and generateResonanceIsomers
# and pick the one with two aromatic rings
smiles = 'C1=CC2C=CC=C3C=CC(=C1)C=23'
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
for mol in spe.molecule:
sssr0 = mol.getSmallestSetOfSmallestRings()
aromaticRingNum = 0
for sr0 in sssr0:
sr0mol = Molecule(atoms=sr0)
if isAromaticRing(sr0mol):
aromaticRingNum += 1
if aromaticRingNum == 2:
break
# extract polyring from the molecule
polyring = mol.getDisparateRings()[1][0]
thermoData = ThermoData(
Tdata=([300, 400, 500, 600, 800, 1000, 1500], "K"),
Cpdata=([0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], "J/(mol*K)"),
H298=(0.0, "kJ/mol"),
S298=(0.0, "J/(mol*K)"),
)
self.database._ThermoDatabase__addPolyRingCorrectionThermoDataFromHeuristic(
thermoData, polyring)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(
thermoData)
ringGroupLabels = [ringGroup.label for ringGroup in ringGroups]
polycyclicGroupLabels = [
polycyclicGroup.label for polycyclicGroup in polycyclicGroups
]
self.assertIn('Benzene', ringGroupLabels)
self.assertIn('Cyclopentene', ringGroupLabels)
self.assertIn('s2_5_6_indene', polycyclicGroupLabels)
self.assertIn('s2_6_6_naphthalene', polycyclicGroupLabels)
def testSymmetryNumberGeneration(self):
"""
Test we generate symmetry numbers correctly.
This uses the new thermo database to generate the H298, used
to select the stablest resonance isomer.
"""
for smiles, symm, H298, S298, Cp300, Cp400, Cp500, Cp600, Cp800, Cp1000, Cp1500 in self.testCases:
molecule = Molecule(SMILES=smiles)
species = Species(molecule=molecule)
species.generateResonanceIsomers()
thermoData = self.database.getThermoDataFromGroups(
Species(molecule=[species.molecule[0]]))
# pick the molecule with lowest H298
molecule = species.molecule[0]
for mol in species.molecule[1:]:
thermoData0 = self.database.getAllThermoData(
Species(molecule=[mol]))[0][0]
for data in self.database.getAllThermoData(
Species(molecule=[mol]))[1:]:
if data.getEnthalpy(298) < thermoData0.getEnthalpy(298):
thermoData0 = data
if thermoData0.getEnthalpy(298) < thermoData.getEnthalpy(298):
thermoData = thermoData0
molecule = mol
self.assertEqual(
molecule.calculateSymmetryNumber(),
symm,
msg="Symmetry number error for {0}".format(smiles))
def testComputeGroupAdditivityThermoForTwoRingMolecule(self):
"""
The molecule being tested has two rings, one is 13cyclohexadiene5methylene
the other is benzene ring. This method is to test thermo estimation will
give two different corrections accordingly.
"""
spec = Species().fromSMILES('CCCCCCCCCCCC(CC=C1C=CC=CC1)c1ccccc1')
spec.generateResonanceIsomers()
thermo = self.database.getThermoDataFromGroups(spec)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(thermo)
self.assertEqual(len(ringGroups),2)
self.assertEqual(len(polycyclicGroups),0)
expected_matchedRingsLabels = ['13cyclohexadiene5methylene', 'Benzene']
expected_matchedRings = [self.database.groups['ring'].entries[label] for label in expected_matchedRingsLabels]
self.assertEqual(set(ringGroups), set(expected_matchedRings))
def testThermoForMonocyclicAndPolycyclicSameMolecule(self):
"""
Test a molecule that has both a polycyclic and a monocyclic ring in the same molecule
"""
spec = Species().fromSMILES('C(CCC1C2CCC1CC2)CC1CCC1')
spec.generateResonanceIsomers()
thermo = self.database.getThermoDataFromGroups(spec)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(thermo)
self.assertEqual(len(ringGroups),1)
self.assertEqual(len(polycyclicGroups),1)
expected_matchedRingsLabels = ['Cyclobutane']
expected_matchedRings = [self.database.groups['ring'].entries[label] for label in expected_matchedRingsLabels]
self.assertEqual(set(ringGroups), set(expected_matchedRings))
expected_matchedPolyringsLabels = ['s3_5_5_ane']
expected_matchedPolyrings = [self.database.groups['polycyclic'].entries[label] for label in expected_matchedPolyringsLabels]
self.assertEqual(set(polycyclicGroups), set(expected_matchedPolyrings))
def testAddPolyRingCorrectionThermoDataFromHeuristicUsingAromaticTricyclic(self):
# create testing molecule
#
# creating it seems not natural in RMG, that's because
# RMG cannot parse the adjacencyList of that isomer correctly
# so here we start with kekulized version and generateResonanceIsomers
# and pick the one with two aromatic rings
smiles = 'C1=CC2C=CC=C3C=CC(=C1)C=23'
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
for mol in spe.molecule:
sssr0 = mol.getSmallestSetOfSmallestRings()
aromaticRingNum = 0
for sr0 in sssr0:
sr0mol = Molecule(atoms=sr0)
if isAromaticRing(sr0mol):
aromaticRingNum += 1
if aromaticRingNum == 2:
break
# extract polyring from the molecule
polyring = mol.getDisparateRings()[1][0]
thermoData = ThermoData(
Tdata = ([300,400,500,600,800,1000,1500],"K"),
Cpdata = ([0.0,0.0,0.0,0.0,0.0,0.0,0.0],"J/(mol*K)"),
H298 = (0.0,"kJ/mol"),
S298 = (0.0,"J/(mol*K)"),
)
self.database._ThermoDatabase__addPolyRingCorrectionThermoDataFromHeuristic(
thermoData, polyring)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(thermoData)
ringGroupLabels = [ringGroup.label for ringGroup in ringGroups]
polycyclicGroupLabels = [polycyclicGroup.label for polycyclicGroup in polycyclicGroups]
self.assertIn('Benzene', ringGroupLabels)
self.assertIn('Cyclopentene', ringGroupLabels)
self.assertIn('s2_5_6_indene', polycyclicGroupLabels)
self.assertIn('s2_6_6_naphthalene', polycyclicGroupLabels)
def testFindAromaticBondsFromSubMolecule(self):
smiles = "C1=CC=C2C=CC=CC2=C1"
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
mol = spe.molecule[1]
# get two SSSRs
SSSR = mol.getSmallestSetOfSmallestRings()
ring1 = SSSR[0]
ring2 = SSSR[1]
# create two testing submols
submol1 = Molecule(atoms=ring1)
submol2 = Molecule(atoms=ring2)
# check with expected results
self.assertEqual(len(findAromaticBondsFromSubMolecule(submol1)), 6)
self.assertEqual(len(findAromaticBondsFromSubMolecule(submol2)), 6)
def testFindAromaticBondsFromSubMolecule(self):
smiles = "C1=CC=C2C=CC=CC2=C1"
spe = Species().fromSMILES(smiles)
spe.generateResonanceIsomers()
mol = spe.molecule[1]
# get two SSSRs
SSSR = mol.getSmallestSetOfSmallestRings()
ring1 = SSSR[0]
ring2 = SSSR[1]
# create two testing submols
submol1 = Molecule(atoms=ring1)
submol2 = Molecule(atoms=ring2)
# check with expected results
self.assertEqual(len(findAromaticBondsFromSubMolecule(submol1)), 6)
self.assertEqual(len(findAromaticBondsFromSubMolecule(submol2)), 6)
def generate():
"""
Test that reaction generation from the available families works.
"""
load()
spcA = Species().fromSMILES('[OH]')
spcs = [Species().fromSMILES('CC'), Species().fromSMILES('[CH3]')]
spcTuples = [(spcA, spc) for spc in spcs]
procnum = 2
reactionList = list(
itertools.chain.from_iterable(react(spcTuples, procnum)))
if not reactionList: return False
for rxn in reactionList:
if not isinstance(rxn, TemplateReaction): return False
return True
def kinetics(label, kinetics):
reactants, products = label.split('_')
reactants = reactants.split('+')
products = products.split('+')
reaction = Reaction(reactants=[], products=[], reversible=True)
for in_list, out_list in [(reactants, reaction.reactants), (products, reaction.products)]:
for i, smiles in enumerate(in_list):
if smiles not in species_dict:
species = Species().fromSMILES(smiles)
species_dict[smiles] = species
species.label = '{0}({1})'.format(species.toChemkin(), len(species_dict))
out_list.append(species_dict[smiles])
reaction.kinetics = kinetics
print repr(reaction)
display(reaction)
addReactionToKineticsLibrary(reaction)
def testComputeGroupAdditivityThermoForTwoRingMolecule(self):
"""
The molecule being tested has two rings, one is 13cyclohexadiene5methylene
the other is benzene ring. This method is to test thermo estimation will
give two different corrections accordingly.
"""
spec = Species().fromSMILES('CCCCCCCCCCCC(CC=C1C=CC=CC1)c1ccccc1')
spec.generateResonanceIsomers()
thermo = self.database.getThermoDataFromGroups(spec)
ringGroups, polycyclicGroups = self.database.getRingGroupsFromComments(
thermo)
self.assertEqual(len(ringGroups), 2)
self.assertEqual(len(polycyclicGroups), 0)
expected_matchedRingsLabels = ['13cyclohexadiene5methylene', 'Benzene']
expected_matchedRings = [
self.database.groups['ring'].entries[label]
for label in expected_matchedRingsLabels
]
self.assertEqual(set(ringGroups), set(expected_matchedRings))
def testSymmetryContributionRadicals(self):
"""
Test that the symmetry contribution is correctly added for radicals
estimated via the HBI method.
"""
spc = Species(molecule=[Molecule().fromSMILES('[CH3]')])
thermoData_lib = self.database.getThermoDataFromLibraries(spc)[0]
thermoData_ga = self.database.getThermoDataFromGroups(spc)
self.assertAlmostEqual(thermoData_lib.getEntropy(298.),
thermoData_ga.getEntropy(298.), 0)
def testSymmetryNumberGeneration(self):
"""
Test we generate symmetry numbers correctly.
This uses the new thermo database to generate the H298, used
to select the stablest resonance isomer.
"""
for smiles, symm, H298, S298, Cp300, Cp400, Cp500, Cp600, Cp800, Cp1000, Cp1500 in self.testCases:
molecule=Molecule(SMILES=smiles)
species = Species(molecule=molecule)
species.generateResonanceIsomers()
thermoData = self.database.getThermoDataFromGroups(Species(molecule=[species.molecule[0]]))
# pick the molecule with lowest H298
molecule = species.molecule[0]
for mol in species.molecule[1:]:
thermoData0 = self.database.getAllThermoData(Species(molecule=[mol]))[0][0]
for data in self.database.getAllThermoData(Species(molecule=[mol]))[1:]:
if data.getEnthalpy(298) < thermoData0.getEnthalpy(298):
thermoData0 = data
if thermoData0.getEnthalpy(298) < thermoData.getEnthalpy(298):
thermoData = thermoData0
molecule = mol
self.assertEqual(molecule.calculateSymmetryNumber(), symm, msg="Symmetry number error for {0}".format(smiles))
def testNewThermoGeneration(self):
"""
Test that the new ThermoDatabase generates appropriate thermo data.
"""
for smiles, symm, H298, S298, Cp300, Cp400, Cp500, Cp600, Cp800, Cp1000, Cp1500 in self.testCases:
Cplist = [Cp300, Cp400, Cp500, Cp600, Cp800, Cp1000, Cp1500]
species = Species().fromSMILES(smiles)
species.generateResonanceIsomers()
thermoData = self.database.getThermoDataFromGroups(species)
molecule = species.molecule[0]
for mol in species.molecule[1:]:
thermoData0 = self.database.getAllThermoData(
Species(molecule=[mol]))[0][0]
for data in self.database.getAllThermoData(
Species(molecule=[mol]))[1:]:
if data[0].getEnthalpy(298) < thermoData0.getEnthalpy(298):
thermoData0 = data[0]
if thermoData0.getEnthalpy(298) < thermoData.getEnthalpy(298):
thermoData = thermoData0
molecule = mol
self.assertAlmostEqual(
H298,
thermoData.getEnthalpy(298) / 4184,
places=1,
msg="H298 error for {0}. Expected {1}, but calculated {2}.".
format(smiles, H298,
thermoData.getEnthalpy(298) / 4184))
self.assertAlmostEqual(
S298,
thermoData.getEntropy(298) / 4.184,
places=1,
msg="S298 error for {0}. Expected {1}, but calculated {2}.".
format(smiles, S298,
thermoData.getEntropy(298) / 4.184))
for T, Cp in zip(self.Tlist, Cplist):
self.assertAlmostEqual(
Cp,
thermoData.getHeatCapacity(T) / 4.184,
places=1,
msg="Cp{3} error for {0}. Expected {1} but calculated {2}."
.format(smiles, Cp,
thermoData.getHeatCapacity(T) / 4.184, T))
delimiter = '\t'
JtoCal = 1.0/4.184
rmg = RMG()
rmg.database = RMGDatabase()
path = os.path.join(settings['database.directory'])
rmg.database.loadThermo(os.path.join(path,'thermo'))
fw = open('output_thermo.txt','w')
fw.write('Name\tSMILES\tH298\tS298\t300 K\t400 K\t500 K\t600 K\t800 K\t1000 K\t1500 K\tComment\n')
for mymol in pybel.readfile("xyz","testfile.txt"):
bel_mol = mymol.write('smi')
mol_smi = bel_mol.split()[0]
mol_name = bel_mol.split()[1]
print mol_name
spc = Species().fromSMILES(mol_smi)
print type(spc)
print spc.toAdjacencyList()
tdt = rmg.database.thermo.getThermoDataFromGroups(spc)
tempstr = mol_name + delimiter + mol_smi + delimiter + str(tdt.H298.value*JtoCal) + delimiter + str(tdt.S298.value*JtoCal)
tempstr = tempstr + delimiter + str(tdt.Cpdata.value[0]*JtoCal) + delimiter + str(tdt.Cpdata.value[1]*JtoCal) + delimiter + str(tdt.Cpdata.value[2]*JtoCal)
tempstr = tempstr + delimiter + str(tdt.Cpdata.value[3]*JtoCal) + delimiter + str(tdt.Cpdata.value[4]*JtoCal) + delimiter + str(tdt.Cpdata.value[5]*JtoCal)
tempstr = tempstr + delimiter + str(tdt.Cpdata.value[6]*JtoCal) + delimiter + tdt.comment + '\n'
fw.write( tempstr )
fw.close()
reactants=reaction.reactants,
products=reaction.products,
reversible=reaction.reversible,
kinetics=reaction.kinetics,
library=libraryName,
)
reactionList.append(library_reaction)
speciesList = []
index = 0
speciesDict = kineticLibrary.getSpecies(
os.path.join(settings["database.directory"], "kinetics", "libraries", libraryName, "dictionary.txt")
)
for spec in speciesDict.values():
index = index + 1
species = Species(molecule=spec.molecule)
species.getThermoData()
species.index = index
speciesList.append(species)
for reaction in reactionList:
for reactant in reaction.reactants:
for spec in speciesList:
if reactant.isIsomorphic(spec):
reactant.index = spec.index
spec.label = reactant.label
for product in reaction.products:
for spec in speciesList:
if product.isIsomorphic(spec):
product.index = spec.index
spec.label = product.label
def testParseThermoComments(self):
"""
Test that the ThermoDatabase.extractSourceFromComments function works properly
on various thermo comments.
"""
from rmgpy.thermo import NASA, NASAPolynomial
# Pure group additivity thermo
propane = Species(
index=3,
label="Propane",
thermo=NASA(
polynomials=[
NASAPolynomial(coeffs=[
3.05257, 0.0125099, 3.79386e-05, -5.12022e-08,
1.87065e-11, -14454.2, 10.0672
],
Tmin=(100, 'K'),
Tmax=(986.57, 'K')),
NASAPolynomial(coeffs=[
5.91316, 0.0218763, -8.17661e-06, 1.49855e-09,
-1.05991e-13, -16038.9, -8.86555
],
Tmin=(986.57, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
comment=
"""Thermo group additivity estimation: group(Cs-CsCsHH) + gauche(Cs(CsCsRR)) + other(R) + group(Cs-CsHHH) + gauche(Cs(Cs(CsRR)RRR)) + other(R) + group(Cs-CsHHH) + gauche(Cs(Cs(CsRR)RRR)) + other(R)"""
),
molecule=[Molecule(SMILES="CCC")])
source = self.database.extractSourceFromComments(propane)
self.assertTrue(
'GAV' in source,
'Should have found that propane thermo source is GAV.')
self.assertEqual(len(source['GAV']['group']), 2)
self.assertEqual(len(source['GAV']['other']), 1)
self.assertEqual(len(source['GAV']['gauche']), 2)
# Pure library thermo
dipk = Species(index=1,
label="DIPK",
thermo=NASA(polynomials=[
NASAPolynomial(coeffs=[
3.35002, 0.017618, -2.46235e-05, 1.7684e-08,
-4.87962e-12, 35555.7, 5.75335
],
Tmin=(100, 'K'),
Tmax=(888.28, 'K')),
NASAPolynomial(coeffs=[
6.36001, 0.00406378, -1.73509e-06, 5.05949e-10,
-4.49975e-14, 35021, -8.41155
],
Tmin=(888.28, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
comment="""Thermo library: DIPK"""),
molecule=[Molecule(SMILES="CC(C)C(=O)C(C)C")])
source = self.database.extractSourceFromComments(dipk)
self.assertTrue('Library' in source)
# Mixed library and HBI thermo
dipk_rad = Species(
index=4,
label="R_tert",
thermo=NASA(polynomials=[
NASAPolynomial(coeffs=[
2.90061, 0.0298018, -7.06268e-05, 6.9636e-08, -2.42414e-11,
54431, 5.44492
],
Tmin=(100, 'K'),
Tmax=(882.19, 'K')),
NASAPolynomial(coeffs=[
6.70999, 0.000201027, 6.65617e-07, -7.99543e-11,
4.08721e-15, 54238.6, -9.73662
],
Tmin=(882.19, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
comment="""Thermo library: DIPK + radical(C2CJCHO)"""),
molecule=[
Molecule(SMILES="C[C](C)C(=O)C(C)C"),
Molecule(SMILES="CC(C)=C([O])C(C)C")
])
source = self.database.extractSourceFromComments(dipk_rad)
self.assertTrue('Library' in source)
self.assertTrue('GAV' in source)
self.assertEqual(len(source['GAV']['radical']), 1)
# Pure QM thermo
cineole = Species(
index=6,
label="Cineole",
thermo=NASA(polynomials=[
NASAPolynomial(coeffs=[
-0.324129, 0.0619667, 9.71008e-05, -1.60598e-07,
6.28285e-11, -38699.9, 29.3686
],
Tmin=(100, 'K'),
Tmax=(985.52, 'K')),
NASAPolynomial(coeffs=[
20.6043, 0.0586913, -2.22152e-05, 4.19949e-09,
-3.06046e-13, -46791, -91.4152
],
Tmin=(985.52, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
comment="""QM MopacMolPM3 calculation attempt 1"""),
molecule=[Molecule(SMILES="CC12CCC(CC1)C(C)(C)O2")])
source = self.database.extractSourceFromComments(cineole)
self.assertTrue('QM' in source)
# Mixed QM and HBI thermo
cineole_rad = Species(
index=7,
label="CineoleRad",
thermo=NASA(
polynomials=[
NASAPolynomial(coeffs=[
-0.2897, 0.0627717, 8.63299e-05, -1.47868e-07,
5.81665e-11, -14017.6, 31.0266
],
Tmin=(100, 'K'),
Tmax=(988.76, 'K')),
NASAPolynomial(coeffs=[
20.4836, 0.0562555, -2.13903e-05, 4.05725e-09,
-2.96023e-13, -21915, -88.1205
],
Tmin=(988.76, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
comment=
"""QM MopacMolPM3 calculation attempt 1 + radical(Cs_P)"""),
molecule=[Molecule(SMILES="[CH2]C12CCC(CC1)C(C)(C)O2")])
source = self.database.extractSourceFromComments(cineole_rad)
self.assertTrue('QM' in source)
self.assertTrue('GAV' in source)
self.assertEqual(len(source['GAV']['radical']), 1)
# No thermo comments
other = Species(
index=7,
label="CineoleRad",
thermo=NASA(
polynomials=[
NASAPolynomial(coeffs=[
-0.2897, 0.0627717, 8.63299e-05, -1.47868e-07,
5.81665e-11, -14017.6, 31.0266
],
Tmin=(100, 'K'),
Tmax=(988.76, 'K')),
NASAPolynomial(coeffs=[
20.4836, 0.0562555, -2.13903e-05, 4.05725e-09,
-2.96023e-13, -21915, -88.1205
],
Tmin=(988.76, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
),
molecule=[Molecule(SMILES="[CH2]C12CCC(CC1)C(C)(C)O2")])
# Function should complain if there's no thermo comments
self.assertRaises(self.database.extractSourceFromComments(cineole_rad))
# Check a dummy species that has plus and minus thermo group contributions
polycyclic = Species(
index=7,
label="dummy",
thermo=NASA(
polynomials=[
NASAPolynomial(coeffs=[
-0.2897, 0.0627717, 8.63299e-05, -1.47868e-07,
5.81665e-11, -14017.6, 31.0266
],
Tmin=(100, 'K'),
Tmax=(988.76, 'K')),
NASAPolynomial(coeffs=[
20.4836, 0.0562555, -2.13903e-05, 4.05725e-09,
-2.96023e-13, -21915, -88.1205
],
Tmin=(988.76, 'K'),
Tmax=(5000, 'K'))
],
Tmin=(100, 'K'),
Tmax=(5000, 'K'),
comment=
"""Thermo group additivity estimation: group(Cs-CsCsHH) + group(Cs-CsCsHH) - ring(Benzene)"""
),
molecule=[Molecule(SMILES="[CH2]C12CCC(CC1)C(C)(C)O2")])
source = self.database.extractSourceFromComments(polycyclic)
self.assertTrue('GAV' in source)
self.assertEqual(source['GAV']['ring'][0][1],
-1) # the weight of benzene contribution should be -1
self.assertEqual(
source['GAV']['group'][0][1],
2) # weight of the group(Cs-CsCsHH) conbtribution should be 2
library_reaction = LibraryReaction(index=reaction.index,
reactants=reaction.reactants,
products=reaction.products,
reversible=reaction.reversible,
kinetics=reaction.kinetics,
library=library_name)
reaction_list.append(library_reaction)
species_list = []
index = 0
species_dict = kinetic_library.get_species(
os.path.join(settings['database.directory'], 'kinetics', 'libraries',
library_name, 'dictionary.txt'))
for spec in list(species_dict.values()):
index = index + 1
species = Species(molecule=spec.molecule)
species.get_thermo_data()
species.index = index
species_list.append(species)
for reaction in reaction_list:
for reactant in reaction.reactants:
for spec in species_list:
if reactant.is_isomorphic(spec):
reactant.index = spec.index
spec.label = reactant.label
for product in reaction.products:
for spec in species_list:
if product.is_isomorphic(spec):
product.index = spec.index
spec.label = product.label
library_reaction = LibraryReaction(index=reaction.index,
reactants=reaction.reactants,
products=reaction.products,
reversible=reaction.reversible,
kinetics=reaction.kinetics,
library=libraryName)
reactionList.append(library_reaction)
speciesList = []
index = 0
speciesDict = kineticLibrary.getSpecies(
os.path.join(settings['database.directory'], 'kinetics', 'libraries',
libraryName, 'dictionary.txt'))
for spec in speciesDict.values():
index = index + 1
species = Species(molecule=spec.molecule)
species.getThermoData()
species.index = index
speciesList.append(species)
for reaction in reactionList:
for reactant in reaction.reactants:
for spec in speciesList:
if reactant.isIsomorphic(spec):
reactant.index = spec.index
spec.label = reactant.label
for product in reaction.products:
for spec in speciesList:
if product.isIsomorphic(spec):
product.index = spec.index
spec.label = product.label
kineticLibrary = database.kinetics.libraries[libraryName]
reactionList = []
for index, entry in kineticLibrary.entries.iteritems():
reaction = entry.item
reaction.kinetics = entry.data
reactionList.append(reaction)
speciesList = []
index = 0
speciesDict = kineticLibrary.getSpecies(
os.path.join(settings['database.directory'], 'kinetics', 'libraries',
libraryName, 'dictionary.txt'))
for spec in speciesDict.values():
index = index + 1
species = Species(molecule=spec.molecule)
species.generateThermoData(database)
species.index = index
speciesList.append(species)
for reaction in reactionList:
for reactant in reaction.reactants:
for spec in speciesList:
if reactant.isIsomorphic(spec):
reactant.index = spec.index
spec.label = reactant.label
for product in reaction.products:
for spec in speciesList:
if product.isIsomorphic(spec):
product.index = spec.index
spec.label = product.label
print 'Loading {0} library'.format(libraryName)
kineticLibrary = database.kinetics.libraries[libraryName]
reactionList = []
for index, entry in kineticLibrary.entries.iteritems():
reaction = entry.item
reaction.kinetics = entry.data
reactionList.append(reaction)
speciesList = []
index = 0
for spec in kineticLibrary.getSpecies().values():
index = index + 1
species = Species(molecule = spec.molecule)
species.generateThermoData(database)
species.index = index
speciesList.append(species)
for reaction in reactionList:
for reactant in reaction.reactants:
for spec in speciesList:
if reactant.isIsomorphic(spec):
reactant.index = spec.index
spec.label = reactant.label
for product in reaction.products:
for spec in speciesList:
if product.isIsomorphic(spec):
product.index = spec.index
spec.label = product.label