def generate(self, target=[], ratios=np.linspace(0, 3, 500)):
if target != []:
target = preprocessing.process_smiles(target)
for i in range(0, len(target)):
self.X_test[i] = target[i]
x_latent = self.smiles_to_latent_model.predict(self.X_test)
molecules = []
smiles_arr = []
for i in range(0, len(target) - 1):
latent1 = x_latent[i:i + 1]
latent0 = x_latent[i + 2:i + 3]
for r in ratios:
rlatent = (1.0 - r) * latent0 + r * latent1
smiles = self.latent_to_smiles(rlatent)
mol = Chem.MolFromSmiles(smiles)
if mol and ((smiles in smiles_arr) == False):
print(smiles)
molecules.append(molecule(smiles))
smiles_arr.append(smiles)
return molecules
def gjfTodat(self, directory, fileName):
atomsNum = 0
lineNum = -1
atom_start = 0
connect_start = 0
geom = ''
multi_done = 0
atom_done = 0
connect_done = 0
fr = file(directory + '/' + fileName, 'r')
tmp_lines = fr.readlines()
for tmp_line in tmp_lines:
lineNum += 1
if multi_done != 1:
tmp_m = self.pattern_multi.match(tmp_line)
if tmp_m:
atom_start = lineNum + 1
multi_done = 1
elif atom_done != 1:
tmp_m = self.pattern_atom.match(tmp_line)
if tmp_m:
atomsNum += 1
else:
atom_done = 1
elif connect_done != 1:
tmp_m = self.pattern_connnect.match(tmp_line)
if tmp_m:
tmp_num = int(tmp_m.group(1))
if (connect_done + tmp_num) == 1:
connect_done -= 1
if tmp_num == 1:
connect_start = lineNum
if tmp_num == atomsNum:
connect_done = 1
else:
connect_done = 0
if connect_start != 0:
# print 'connectivity info extracted successfully!'
pass
else:
print 'connectivity info extracted not successfully!'
geom = geometryExtractor.mominertGeometryExtractor(tmp_lines[atom_start: atom_start + atomsNum])
molecule1 = chem.molecule(geom=tmp_lines[atom_start: atom_start + atomsNum], connect=tmp_lines[connect_start: connect_start + atomsNum], atomsNum=atomsNum)
if __OOSYSTEM__ == True:
molecule1.fulfillBonds()
# molecule1.displayBonds()
rotations = molecule1.getRotations()
fr.close()
fw = file(directory + '/' + fileName[7:-4] + '.dat', 'w')
fw.write(fileName[7:-4] + '\n' + 'ANGS\n' + str(atomsNum) + '\n' + geom)
fw.write(''.join(tmp_rot.singleGroupInfo() for tmp_rot in rotations))
# for tmp_rot in rotations:
# fw.write(tmp_rot.group1Info())
fw.write('0 0\n\n\n\n\n\n\n')
fw.close()
return fileName[7:-4] + '.dat'
def generate(self, target=[], hit_rate=100, preprocessing_instance=None):
if target != []:
target = preprocessing_instance.process_smiles(target)
for i in range(0, len(target)):
self.X_test[i] = target[i]
x_latent = self.smiles_to_latent_model.predict(self.X_test)
molecules = []
smiles_arr = []
for i in range(0, len(target)-1):
latent1 = x_latent[i:i+1]
latent0 = x_latent[i+2:i+3]
for r in np.linspace(0,3,hit_rate):
rlatent = (1.0-r)*latent0 + r*latent1
smiles = self.latent_to_smiles(rlatent, preprocessing_instance)
mol = Chem.MolFromSmiles(smiles)
if mol and ((smiles in smiles_arr) == False):
print(smiles, "adding to array!", "\n\n")
molecules.append(molecule(smiles))
smiles_arr.append(smiles)
return molecules
def readConformers(self, fileName, path=''):
# varibles
molecules = []
# flags
energyStart_done = 0
energy_done = 0
atomStart_done = 0
coordinate_done = 0
# temporary variables
tmp_energy = 0.0
tmp_geom = []
tmp_coordinate = ''
atomStartLine = 0
fr = file(os.path.join(path, fileName), 'r')
tmp_lines = fr.readlines()
for (index, tmp_line) in enumerate(tmp_lines):
if atomStart_done != 1:
tmp_m = Balloon.pattern_sdfAtomStart.match(tmp_line)
if tmp_m:
atomStartLine = index + 1
atomStart_done = 1
elif coordinate_done != 1:
tmp_m = Balloon.pattern_sdfAtom.match(tmp_line)
if tmp_m:
tmp_coordinate = ''.join([
tmp_m.group(4), ' ',
tmp_m.group(1), ' ',
tmp_m.group(2), ' ',
tmp_m.group(3)
])
tmp_geom.append(tmp_coordinate)
else:
coordinate_done = 1
elif energyStart_done != 1:
tmp_m = Balloon.pattern_sdfEnergyStart.match(tmp_line)
if tmp_m:
energyStart_done = 1
elif energy_done != 1:
tmp_m = Balloon.pattern_sdfEnergy.match(tmp_line)
if tmp_m:
tmp_energy = float(tmp_m.group(1))
tmp_molecule = chem.molecule()
tmp_molecule.getGjfGeom(tmp_geom)
tmp_molecule.setZPE(tmp_energy)
molecules.append(tmp_molecule)
atomStart_done = 0
coordinate_done = 0
energyStart_done = 0
tmp_energy = 0.0
tmp_geom = []
tmp_coordinate = ''
fr.close()
return molecules
def readGjfFile(self, fileName, directory='', moleculeLabel=''): #definition of flags gjfCommand_done = -1 gjfMulti_done = -1 geomDone = -1 #definition of temporary variables tmp_m = [] #match result tmp_multi = 1 tmp_geom = '' lineStart = 0 lineEnd = 0 # print fileName gjfFile = file(os.path.join(directory, fileName), 'r') tmp_lines = gjfFile.readlines() for (lineNum, tmp_line) in enumerate(tmp_lines): if gjfCommand_done != 1: tmp_m = self.pattern_gjfCommand.match(tmp_line) if tmp_m: gjfCommand_done = 1 elif gjfMulti_done != 1: tmp_m = self.pattern_gjfMulti.match(tmp_line) if tmp_m: lineStart = lineNum tmp_multi = int(tmp_m.group(2)) geomDone = 0 gjfMulti_done = 1 elif geomDone != 1: tmp_m = self.pattern_blankLine.match(tmp_line) if tmp_m: lineEnd = lineNum tmp_geom = tmp_lines[lineStart+1: lineEnd] geomDone = 1 if geomDone != 1: print 'Sorry! The input file is not a standard gjf file!' else: pass # print 'Gjf file read in successfully!' gjfFile.close() self.mole = chem.molecule() self.mole.getGjfGeom(tmp_geom) self.mole.setSpinMultiplicity(tmp_multi) self.mole.calcFormula() if moleculeLabel == '': self.mole.setLabel(self.mole.formula) else: self.mole.setLabel(moleculeLabel) self.mole.fulfillBonds() return self.mole
def readGjfFile(self, fileName, directory='', moleculeLabel=''):
#definition of flags
gjfCommand_done = -1
gjfMulti_done = -1
geomDone = -1
#definition of temporary variables
tmp_m = [] #match result
tmp_multi = 1
tmp_geom = ''
lineStart = 0
lineEnd = 0
# print fileName
gjfFile = file(os.path.join(directory, fileName), 'r')
tmp_lines = gjfFile.readlines()
for (lineNum, tmp_line) in enumerate(tmp_lines):
if gjfCommand_done != 1:
tmp_m = self.pattern_gjfCommand.match(tmp_line)
if tmp_m:
gjfCommand_done = 1
elif gjfMulti_done != 1:
tmp_m = self.pattern_gjfMulti.match(tmp_line)
if tmp_m:
lineStart = lineNum
tmp_multi = int(tmp_m.group(2))
geomDone = 0
gjfMulti_done = 1
elif geomDone != 1:
tmp_m = self.pattern_blankLine.match(tmp_line)
if tmp_m:
lineEnd = lineNum
tmp_geom = tmp_lines[lineStart + 1:lineEnd]
geomDone = 1
if geomDone != 1:
print 'Sorry! The input file is not a standard gjf file!'
else:
pass
# print 'Gjf file read in successfully!'
gjfFile.close()
self.mole = chem.molecule()
self.mole.getGjfGeom(tmp_geom)
self.mole.setSpinMultiplicity(tmp_multi)
self.mole.calcFormula()
if moleculeLabel == '':
self.mole.setLabel(self.mole.formula)
else:
self.mole.setLabel(moleculeLabel)
self.mole.fulfillBonds()
return self.mole
def readConformers(self, fileName, path=''): # varibles molecules = [] # flags energyStart_done = 0 energy_done = 0 atomStart_done = 0 coordinate_done = 0 # temporary variables tmp_energy = 0.0 tmp_geom = [] tmp_coordinate = '' atomStartLine = 0 fr = file(os.path.join(path, fileName), 'r') tmp_lines = fr.readlines() for (index, tmp_line) in enumerate(tmp_lines): if atomStart_done != 1: tmp_m = Balloon.pattern_sdfAtomStart.match(tmp_line) if tmp_m: atomStartLine = index + 1 atomStart_done = 1 elif coordinate_done != 1: tmp_m = Balloon.pattern_sdfAtom.match(tmp_line) if tmp_m: tmp_coordinate = ''.join([tmp_m.group(4), ' ', tmp_m.group(1), ' ', tmp_m.group(2), ' ', tmp_m.group(3)]) tmp_geom.append(tmp_coordinate) else: coordinate_done = 1 elif energyStart_done != 1: tmp_m = Balloon.pattern_sdfEnergyStart.match(tmp_line) if tmp_m: energyStart_done = 1 elif energy_done != 1: tmp_m = Balloon.pattern_sdfEnergy.match(tmp_line) if tmp_m: tmp_energy = float(tmp_m.group(1)) tmp_molecule = chem.molecule() tmp_molecule.getGjfGeom(tmp_geom) tmp_molecule.setZPE(tmp_energy) molecules.append(tmp_molecule) atomStart_done = 0 coordinate_done = 0 energyStart_done = 0 tmp_energy = 0.0 tmp_geom = [] tmp_coordinate = '' fr.close() return molecules
def readGjfGeom(self, gjfGeom, moleculeLabel='', multiplicity=1):
tmp_multi = multiplicity
tmp_geom = gjfGeom.strip()
tmp_geom = tmp_geom.split('\n')
self.mole = chem.molecule()
self.mole.getGjfGeom(tmp_geom)
self.mole.setSpinMultiplicity(tmp_multi)
self.mole.calcFormula()
if moleculeLabel == '':
self.mole.setLabel(self.mole.formula)
else:
self.mole.setLabel(moleculeLabel)
self.mole.fulfillBonds()
return self.mole
def readGjfGeom(self, gjfGeom, moleculeLabel='', multiplicity=1):
tmp_multi = multiplicity
tmp_geom = gjfGeom.strip()
tmp_geom = tmp_geom.split('\n')
self.mole = chem.molecule()
self.mole.getGjfGeom(tmp_geom)
self.mole.setSpinMultiplicity(tmp_multi)
self.mole.calcFormula()
if moleculeLabel == '':
self.mole.setLabel(self.mole.formula)
else:
self.mole.setLabel(moleculeLabel)
self.mole.fulfillBonds()
return self.mole
def readGjfGeom(self, gjfGeom, moleculeLabel='', multiplicity=1):
tmp_multi = multiplicity
tmp_geom = gjfGeom.strip()
tmp_geom = tmp_geom.split('\n')
self.mole = chem.molecule()
try:
self.mole.getGjfGeom(tmp_geom)
except:
print 'Sorry! The input file is not in the expected format!'
raise common_api.readGjfGeomError
self.mole.setSpinMultiplicity(tmp_multi)
self.mole.calcFormula()
tmp_m = self.pattern_formula.match(self.mole.formula)
if tmp_m:
if tmp_m.group(1) == '':
carbonNumber = 1
else:
carbonNumber = int(tmp_m.group(1))
if carbonNumber < 3:
print 'Please submit a species with carbon atoms >= 3!'
raise common_api.carbonLessThan3Error
else:
print 'Sorry! Not an alkane or alkene or their radical! Not supported temporarily!'
raise common_api.beyondSpeciesRangeError
if moleculeLabel == '':
self.mole.setLabel(self.mole.formula)
else:
self.mole.setLabel(moleculeLabel)
try:
self.mole.fulfillBonds()
except:
print 'mole.fulfillBonds() failed!'
raise common_api.readGjfGeomError
if self.mole.existRings():
print 'Sorry! Structure with rings is not supported temporarily!'
raise common_api.ringExistingError
return self.mole
def genFrogInputFromGjf(self, fileList, path='', jobName=''):
for tmp_file in fileList:
gjfCommand_done = -1
gjfMulti_done = -1
geomDone = -1
lineStart = 0
lineEnd = 0
if re.search('[Tt][sS]', tmp_file):
self.setTS(True)
else:
self.setTS(False)
if jobName == '':
tmp_dir = tmp_file[0:-4] + '_1_confSearch'
else:
tmp_dir = jobName
if path == '':
tmp_dir_path = tmp_dir
fr = file(tmp_file, 'r')
else:
tmp_dir_path = os.path.join(path, tmp_dir)
fr = file(os.path.join(path, tmp_file), 'r')
tmp_lines = fr.readlines()
for (lineNum, tmp_line) in enumerate(tmp_lines):
if gjfCommand_done != 1:
tmp_m = pattern_gjfCommand.match(tmp_line)
if tmp_m:
gjfCommand_done = 1
elif gjfMulti_done != 1:
tmp_m = pattern_gjfMulti.match(tmp_line)
if tmp_m:
lineStart = lineNum
multi = int(tmp_m.group(2))
geomDone = 0
gjfMulti_done = 1
elif geomDone != 1:
tmp_m = pattern_blankLine.match(tmp_line)
if tmp_m:
lineEnd = lineNum
geomDone = 1
if os.path.exists(tmp_dir):
shutil.rmtree(tmp_dir)
os.mkdir(tmp_dir)
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.gjf'), 'w')
fw.write(
'''%mem=28GB
%nprocshared=12
''')
if self._TS == False:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt freq')
else:
fw.write('#p b3lyp/cbsb7 opt freq')
else:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt=(TS, calcfc) freq')
else:
fw.write('#p b3lyp/cbsb7 opt=(TS, calcfc) freq')
if self._dispersionD3 == False:
fw.write('\n')
else:
fw.write(' EmpiricalDispersion=GD3\n')
fw.write('''
using ub3lyp/6-31G(d) to scan
''')
fw.write(''.join(tmp_lines[lineStart: lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + fw.name + ' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.xyz'), 'w')
fw.write(str(lineEnd - lineStart - 1) + '\n')
fw.write(tmp_file[0:-4] + '\n')
fw.write(''.join(tmp_lines[lineStart+1: lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + fw.name + ' > log_dos2unix.txt 2>&1')
os.system('E:\\softwares\\OpenBabel-2.3.72\\babel.exe -ixyz ' + os.path.join(tmp_dir_path, tmp_dir+'.xyz') + ' -osdf ' + os.path.join(tmp_dir_path, tmp_dir+'.sdf') + ' > log_dos2unix.txt 2>&1')
openBabelWrong = False
tmp_molecule = chem.molecule(geom=tmp_lines[lineStart+1: lineEnd])
tmp_molecule.fulfillBonds()
if os.path.exists(os.path.join(tmp_dir_path, tmp_dir+'.sdf')):
fr = file(os.path.join(tmp_dir_path, tmp_dir+'.sdf'), 'r')
tmp2_lines = fr.readlines()
fr.close()
tmp_num = map(int, tmp2_lines[3].split()[0:2])
if tmp_num[1] < len(tmp_molecule.bonds):
openBabelWrong = True
elif tmp_num[1] > len(tmp_molecule.bonds):
print 'Error! Open babel bond number > len(tmp_molecule.bonds)', tmp_dir
else:
openBabelWrong = True
if openBabelWrong:
print 'Warning! Open babel transformation bug! Chem used to regenerate the bonds!', tmp_dir
tmp_molecule.generateSDFFile(directory=tmp_dir_path, fileName=tmp_dir+'.sdf', moleculeLabel=tmp_file[0:-4])
# fw = file(os.path.join(tmp_dir_path, tmp_dir+'.sdf'), 'w')
# tmp2_lines[3] = ''.join([' ', '%2d'%tmp_num[0], ' ', '%2d'%(len(tmp_molecule.bonds)), tmp2_lines[3][6:]])
# fw.writelines(tmp2_lines[0:3+tmp_num[0]+1])
# for tmp_bond in tmp_molecule.bonds:
# fw.write(''.join([' ', '%2d'%tmp_bond.atom1.label, ' ', '%2d'%tmp_bond.atom2.label, ' ', '%2d'%tmp_bond.bondOrder, ' 0 0 0 0\n']))
# fw.write(
# '''M END
# $$$$
# ''')
# fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + os.path.join(tmp_dir_path, tmp_dir+'.sdf') + ' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.job'), 'w')
fw.write(
# cce cluster
'''#!/bin/sh
cd ''' + self.jobLocation + '/' + tmp_dir + '''
python /home/apps/Frog2/www_iMolecule.py -osmi ''' + tmp_dir + '''.smiles -logFile ''' + tmp_dir + '''.log -ounsolved Unsolved.data -wrkPath . -eini 100.0 -mcsteps 100 -emax 50 -i3Dsdf ''' + tmp_dir + '''.sdf -osdf out_''' + tmp_dir + '''.sdf -unambiguate -mini -multi 250 &>> log_''' + tmp_dir + '''.txt
''')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + fw.name + ' > log_dos2unix.txt 2>&1')
coordinate_done = 0
if coordinate_done != 1:
tmp_m = pattern_endline.match(tmp_line)
if tmp_m:
if i > geom_start:
geom_end = i
coordinate_done = 1
tmp_m = pattern_energy.match(tmp_line)
if tmp_m:
tmp_energy = float(tmp_m.group(1))
tmp_m = pattern_optimized.match(tmp_line)
if tmp_m:
energy.append(tmp_energy)
if (geom_end - geom_start) != atomsNum:
print 'Error! The number of atoms is not correct!'
tmp_mole = chem.molecule()
tmp_mole.getLogGeom(
tmp_lines[geom_start:geom_start + atomsNum])
tmp_mole.changeLabel(tmp_file2[0:-4])
geoms.append(tmp_mole)
dihedral_done = 0
standard_done = 1
coordinate_done = 1
energy_done = 1
optimized_done = 1
elif dihedral_done != 1:
tmp_m = pattern_dihedral.match(tmp_line)
if tmp_m:
tmp_dihedral = float(tmp_m.group(1))
if len(dihedral) > 0:
if tmp_dihedral < dihedral[-1]:
def genBalloonInputFromGjf(self, fileList, path='', jobName=''):
for tmp_file in fileList:
gjfCommand_done = -1
gjfMulti_done = -1
geomDone = -1
lineStart = 0
lineEnd = 0
if re.search('[Tt][sS]', tmp_file):
self.setTS(True)
else:
self.setTS(False)
if jobName == '':
tmp_dir = tmp_file[0:-4] + '_1_confSearch'
else:
tmp_dir = jobName
if path == '':
tmp_dir_path = tmp_dir
fr = file(tmp_file, 'r')
else:
tmp_dir_path = os.path.join(path, tmp_dir)
fr = file(os.path.join(path, tmp_file), 'r')
tmp_lines = fr.readlines()
for (lineNum, tmp_line) in enumerate(tmp_lines):
if gjfCommand_done != 1:
tmp_m = pattern_gjfCommand.match(tmp_line)
if tmp_m:
gjfCommand_done = 1
elif gjfMulti_done != 1:
tmp_m = pattern_gjfMulti.match(tmp_line)
if tmp_m:
lineStart = lineNum
multi = int(tmp_m.group(2))
geomDone = 0
gjfMulti_done = 1
elif geomDone != 1:
tmp_m = pattern_blankLine.match(tmp_line)
if tmp_m:
lineEnd = lineNum
geomDone = 1
if os.path.exists(tmp_dir):
shutil.rmtree(tmp_dir)
os.mkdir(tmp_dir)
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.gjf'), 'w')
fw.write('''%mem=28GB
%nprocshared=12
''')
if self._TS == False:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt freq')
else:
fw.write('#p b3lyp/cbsb7 opt freq')
else:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt=(TS, calcfc) freq')
else:
fw.write('#p b3lyp/cbsb7 opt=(TS, calcfc) freq')
if self._dispersionD3 == False:
fw.write('\n')
else:
fw.write(' EmpiricalDispersion=GD3\n')
fw.write('''
using ub3lyp/6-31G(d) to scan
''')
fw.write(''.join(tmp_lines[lineStart:lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
fw.name + ' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.xyz'), 'w')
fw.write(str(lineEnd - lineStart - 1) + '\n')
fw.write(tmp_file[0:-4] + '\n')
fw.write(''.join(tmp_lines[lineStart + 1:lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
fw.name + ' > log_dos2unix.txt 2>&1')
os.system('E:\\softwares\\OpenBabel-2.3.72\\babel.exe -ixyz ' +
os.path.join(tmp_dir_path, tmp_dir + '.xyz') +
' -osdf ' +
os.path.join(tmp_dir_path, tmp_dir + '.sdf') +
' > log_dos2unix.txt 2>&1')
fr = file(os.path.join(tmp_dir_path, tmp_dir + '.sdf'), 'r')
tmp2_lines = fr.readlines()
fr.close()
tmp_num = map(int, tmp2_lines[3].split()[0:2])
tmp_molecule = chem.molecule(geom=tmp_lines[lineStart + 1:lineEnd])
tmp_molecule.fulfillBonds()
if tmp_num[1] < len(tmp_molecule.bonds):
print 'Warning! Open babel transformation bug! Chem used to regenerate the bonds!', tmp_dir
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.sdf'), 'w')
tmp2_lines[3] = ''.join([
' ',
'%2d' % tmp_num[0], ' ',
'%2d' % (len(tmp_molecule.bonds)), tmp2_lines[3][6:]
])
fw.writelines(tmp2_lines[0:3 + tmp_num[0] + 1])
for tmp_bond in tmp_molecule.bonds:
fw.write(''.join([
' ',
'%2d' % tmp_bond.atom1.label, ' ',
'%2d' % tmp_bond.atom2.label, ' ',
'%2d' % tmp_bond.bondOrder, ' 0 0 0 0\n'
]))
fw.write('''M END
$$$$
''')
fw.close()
elif tmp_num[1] > len(tmp_molecule.bonds):
print 'Error! Open babel bond number > len(tmp_molecule.bonds)', tmp_dir
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
os.path.join(tmp_dir_path, tmp_dir + '.sdf') +
' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.job'), 'w')
fw.write(
# linux bash
'''#!/bin/sh
cd ''' + self.jobLocation + '/' + tmp_dir + '''
/home/apps/balloon/balloon -f /home/hetanjin/apps/balloon/MMFF94.mff --nconfs 300 --stereo --addConformerNumberToName '''
+ tmp_dir + '''.sdf out_''' + tmp_dir + '''.sdf &>> log_''' +
tmp_dir + '''.txt
''')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
fw.name + ' > log_dos2unix.txt 2>&1')
def genFrogInputFromGjf(self, fileList, path='', jobName=''):
for tmp_file in fileList:
gjfCommand_done = -1
gjfMulti_done = -1
geomDone = -1
lineStart = 0
lineEnd = 0
if re.search('[Tt][sS]', tmp_file):
self.setTS(True)
else:
self.setTS(False)
if jobName == '':
tmp_dir = tmp_file[0:-4] + '_1_confSearch'
else:
tmp_dir = jobName
if path == '':
tmp_dir_path = tmp_dir
fr = file(tmp_file, 'r')
else:
tmp_dir_path = os.path.join(path, tmp_dir)
fr = file(os.path.join(path, tmp_file), 'r')
tmp_lines = fr.readlines()
for (lineNum, tmp_line) in enumerate(tmp_lines):
if gjfCommand_done != 1:
tmp_m = pattern_gjfCommand.match(tmp_line)
if tmp_m:
gjfCommand_done = 1
elif gjfMulti_done != 1:
tmp_m = pattern_gjfMulti.match(tmp_line)
if tmp_m:
lineStart = lineNum
multi = int(tmp_m.group(2))
geomDone = 0
gjfMulti_done = 1
elif geomDone != 1:
tmp_m = pattern_blankLine.match(tmp_line)
if tmp_m:
lineEnd = lineNum
geomDone = 1
if os.path.exists(tmp_dir):
shutil.rmtree(tmp_dir)
os.mkdir(tmp_dir)
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.gjf'), 'w')
fw.write('''%mem=28GB
%nprocshared=12
''')
if self._TS == False:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt freq')
else:
fw.write('#p b3lyp/cbsb7 opt freq')
else:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt=(TS, calcfc) freq')
else:
fw.write('#p b3lyp/cbsb7 opt=(TS, calcfc) freq')
if self._dispersionD3 == False:
fw.write('\n')
else:
fw.write(' EmpiricalDispersion=GD3\n')
fw.write('''
using ub3lyp/6-31G(d) to scan
''')
fw.write(''.join(tmp_lines[lineStart:lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
fw.name + ' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.xyz'), 'w')
fw.write(str(lineEnd - lineStart - 1) + '\n')
fw.write(tmp_file[0:-4] + '\n')
fw.write(''.join(tmp_lines[lineStart + 1:lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
fw.name + ' > log_dos2unix.txt 2>&1')
os.system('E:\\softwares\\OpenBabel-2.3.72\\babel.exe -ixyz ' +
os.path.join(tmp_dir_path, tmp_dir + '.xyz') +
' -osdf ' +
os.path.join(tmp_dir_path, tmp_dir + '.sdf') +
' > log_dos2unix.txt 2>&1')
openBabelWrong = False
tmp_molecule = chem.molecule(geom=tmp_lines[lineStart + 1:lineEnd])
tmp_molecule.fulfillBonds()
if os.path.exists(os.path.join(tmp_dir_path, tmp_dir + '.sdf')):
fr = file(os.path.join(tmp_dir_path, tmp_dir + '.sdf'), 'r')
tmp2_lines = fr.readlines()
fr.close()
tmp_num = map(int, tmp2_lines[3].split()[0:2])
if tmp_num[1] < len(tmp_molecule.bonds):
openBabelWrong = True
elif tmp_num[1] > len(tmp_molecule.bonds):
print 'Error! Open babel bond number > len(tmp_molecule.bonds)', tmp_dir
else:
openBabelWrong = True
if openBabelWrong:
print 'Warning! Open babel transformation bug! Chem used to regenerate the bonds!', tmp_dir
tmp_molecule.generateSDFFile(directory=tmp_dir_path,
fileName=tmp_dir + '.sdf',
moleculeLabel=tmp_file[0:-4])
# fw = file(os.path.join(tmp_dir_path, tmp_dir+'.sdf'), 'w')
# tmp2_lines[3] = ''.join([' ', '%2d'%tmp_num[0], ' ', '%2d'%(len(tmp_molecule.bonds)), tmp2_lines[3][6:]])
# fw.writelines(tmp2_lines[0:3+tmp_num[0]+1])
# for tmp_bond in tmp_molecule.bonds:
# fw.write(''.join([' ', '%2d'%tmp_bond.atom1.label, ' ', '%2d'%tmp_bond.atom2.label, ' ', '%2d'%tmp_bond.bondOrder, ' 0 0 0 0\n']))
# fw.write(
# '''M END
# $$$$
# ''')
# fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
os.path.join(tmp_dir_path, tmp_dir + '.sdf') +
' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir + '.job'), 'w')
fw.write(
# cce cluster
'''#!/bin/sh
cd ''' + self.jobLocation + '/' + tmp_dir + '''
python /home/apps/Frog2/www_iMolecule.py -osmi ''' + tmp_dir +
'''.smiles -logFile ''' + tmp_dir +
'''.log -ounsolved Unsolved.data -wrkPath . -eini 100.0 -mcsteps 100 -emax 50 -i3Dsdf '''
+ tmp_dir + '''.sdf -osdf out_''' + tmp_dir +
'''.sdf -unambiguate -mini -multi 250 &>> log_''' + tmp_dir +
'''.txt
''')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " +
fw.name + ' > log_dos2unix.txt 2>&1')
def gjfTodat(self, directory, fileName):
atomsNum = 0
lineNum = -1
atom_start = 0
connect_start = 0
geom = ''
multi_done = 0
atom_done = 0
connect_done = 0
fr = file(directory + '/' + fileName, 'r')
tmp_lines = fr.readlines()
for tmp_line in tmp_lines:
lineNum += 1
if multi_done != 1:
tmp_m = self.pattern_multi.match(tmp_line)
if tmp_m:
atom_start = lineNum + 1
multi_done = 1
elif atom_done != 1:
tmp_m = self.pattern_atom.match(tmp_line)
if tmp_m:
atomsNum += 1
else:
atom_done = 1
elif connect_done != 1:
tmp_m = self.pattern_connnect.match(tmp_line)
if tmp_m:
tmp_num = int(tmp_m.group(1))
if (connect_done + tmp_num) == 1:
connect_done -= 1
if tmp_num == 1:
connect_start = lineNum
if tmp_num == atomsNum:
connect_done = 1
else:
connect_done = 0
if connect_start != 0:
# print 'connectivity info extracted successfully!'
pass
else:
print 'connectivity info extracted not successfully!'
geom = geometryExtractor.mominertGeometryExtractor(
tmp_lines[atom_start:atom_start + atomsNum])
molecule1 = chem.molecule(
geom=tmp_lines[atom_start:atom_start + atomsNum],
connect=tmp_lines[connect_start:connect_start + atomsNum],
atomsNum=atomsNum)
if __OOSYSTEM__ == True:
molecule1.fulfillBonds()
# molecule1.displayBonds()
rotations = molecule1.getRotations()
fr.close()
fw = file(directory + '/' + fileName[7:-4] + '.dat', 'w')
fw.write(fileName[7:-4] + '\n' + 'ANGS\n' + str(atomsNum) + '\n' +
geom)
fw.write(''.join(tmp_rot.singleGroupInfo() for tmp_rot in rotations))
# for tmp_rot in rotations:
# fw.write(tmp_rot.group1Info())
fw.write('0 0\n\n\n\n\n\n\n')
fw.close()
return fileName[7:-4] + '.dat'
def genBalloonInputFromGjf(self, fileList, path='', jobName=''):
for tmp_file in fileList:
gjfCommand_done = -1
gjfMulti_done = -1
geomDone = -1
lineStart = 0
lineEnd = 0
if re.search('[Tt][sS]', tmp_file):
self.setTS(True)
else:
self.setTS(False)
if jobName == '':
tmp_dir = tmp_file[0:-4] + '_1_confSearch'
else:
tmp_dir = jobName
if path == '':
tmp_dir_path = tmp_dir
fr = file(tmp_file, 'r')
else:
tmp_dir_path = os.path.join(path, tmp_dir)
fr = file(os.path.join(path, tmp_file), 'r')
tmp_lines = fr.readlines()
for (lineNum, tmp_line) in enumerate(tmp_lines):
if gjfCommand_done != 1:
tmp_m = pattern_gjfCommand.match(tmp_line)
if tmp_m:
gjfCommand_done = 1
elif gjfMulti_done != 1:
tmp_m = pattern_gjfMulti.match(tmp_line)
if tmp_m:
lineStart = lineNum
multi = int(tmp_m.group(2))
geomDone = 0
gjfMulti_done = 1
elif geomDone != 1:
tmp_m = pattern_blankLine.match(tmp_line)
if tmp_m:
lineEnd = lineNum
geomDone = 1
if os.path.exists(tmp_dir):
shutil.rmtree(tmp_dir)
os.mkdir(tmp_dir)
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.gjf'), 'w')
fw.write(
'''%mem=28GB
%nprocshared=12
''')
if self._TS == False:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt freq')
else:
fw.write('#p b3lyp/cbsb7 opt freq')
else:
if multi != 1:
fw.write('#p ub3lyp/cbsb7 opt=(TS, calcfc) freq')
else:
fw.write('#p b3lyp/cbsb7 opt=(TS, calcfc) freq')
if self._dispersionD3 == False:
fw.write('\n')
else:
fw.write(' EmpiricalDispersion=GD3\n')
fw.write('''
using ub3lyp/6-31G(d) to scan
''')
fw.write(''.join(tmp_lines[lineStart: lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + fw.name + ' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.xyz'), 'w')
fw.write(str(lineEnd - lineStart - 1) + '\n')
fw.write(tmp_file[0:-4] + '\n')
fw.write(''.join(tmp_lines[lineStart+1: lineEnd]) + '\n\n\n\n\n')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + fw.name + ' > log_dos2unix.txt 2>&1')
os.system('E:\\softwares\\OpenBabel-2.3.72\\babel.exe -ixyz ' + os.path.join(tmp_dir_path, tmp_dir+'.xyz') + ' -osdf ' + os.path.join(tmp_dir_path, tmp_dir+'.sdf') + ' > log_dos2unix.txt 2>&1')
fr = file(os.path.join(tmp_dir_path, tmp_dir+'.sdf'), 'r')
tmp2_lines = fr.readlines()
fr.close()
tmp_num = map(int, tmp2_lines[3].split()[0:2])
tmp_molecule = chem.molecule(geom=tmp_lines[lineStart+1: lineEnd])
tmp_molecule.fulfillBonds()
if tmp_num[1] < len(tmp_molecule.bonds):
print 'Warning! Open babel transformation bug! Chem used to regenerate the bonds!', tmp_dir
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.sdf'), 'w')
tmp2_lines[3] = ''.join([' ', '%2d'%tmp_num[0], ' ', '%2d'%(len(tmp_molecule.bonds)), tmp2_lines[3][6:]])
fw.writelines(tmp2_lines[0:3+tmp_num[0]+1])
for tmp_bond in tmp_molecule.bonds:
fw.write(''.join([' ', '%2d'%tmp_bond.atom1.label, ' ', '%2d'%tmp_bond.atom2.label, ' ', '%2d'%tmp_bond.bondOrder, ' 0 0 0 0\n']))
fw.write(
'''M END
$$$$
''')
fw.close()
elif tmp_num[1] > len(tmp_molecule.bonds):
print 'Error! Open babel bond number > len(tmp_molecule.bonds)', tmp_dir
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + os.path.join(tmp_dir_path, tmp_dir+'.sdf') + ' > log_dos2unix.txt 2>&1')
fw = file(os.path.join(tmp_dir_path, tmp_dir+'.job'), 'w')
fw.write(
# linux bash
'''#!/bin/sh
cd ''' + self.jobLocation + '/' + tmp_dir + '''
/home/apps/balloon/balloon -f /home/hetanjin/apps/balloon/MMFF94.mff --nconfs 300 --stereo --addConformerNumberToName ''' + tmp_dir + '''.sdf out_''' + tmp_dir + '''.sdf &>> log_''' + tmp_dir + '''.txt
''')
fw.close()
os.system("..\\dos2unix-6.0.6-win64\\bin\\dos2unix.exe " + fw.name + ' > log_dos2unix.txt 2>&1')
import xlrd
import re
import numpy as np
import os
import xlsxwriter
import phaseSpaceIntegral
import phys
import chem
import textExtractor
# constant
phys1 = phys.phys()
molecule1 = chem.molecule()
pattern_scanFileName = re.compile('^([CHO0-9]+_[0-9]+)_.*scan.*$')
pattern_geomFileName = re.compile('^([CHO0-9]+_[0-9]+)_.*opt.*$')
pattern_freqCom = re.compile('^.*#[PN]? Geom=AllCheck Guess=TCheck SCRF=Check.*Freq.*$')
# note that Input orientation should be used in MESMER rather than standard orientation when hessian used
pattern_input = re.compile('^.*Input orientation:.*$')
# pattern_standard = re.compile('^.*Standard orientation:.*$')
pattern_endline = re.compile('^.*---------------------------------------------------------------------.*$')
# variables
temperature = [298.15] + range(300,2600,100)
# target variables
HR_dict = {}
mole_dict = {}
# read data and get integral
wbr = xlrd.open_workbook('HR_fit.xls')
def readGjfFile(self, gjfFile, moleculeLabel=''): #definition of flags gjfCommand_done = -1 gjfMulti_done = -1 geomDone = -1 #definition of temporary variables tmp_m = [] #match result tmp_multi = 1 tmp_geom = '' lineStart = 0 lineEnd = 0 BASE_DIR = os.path.dirname(os.path.dirname(__file__)) # print fileName tmp_lines = gjfFile.readlines() for (lineNum, tmp_line) in enumerate(tmp_lines): if gjfCommand_done != 1: tmp_m = self.pattern_gjfCommand.match(tmp_line) if tmp_m: gjfCommand_done = 1 elif gjfMulti_done != 1: tmp_m = self.pattern_gjfMulti.match(tmp_line) if tmp_m: lineStart = lineNum tmp_multi = int(tmp_m.group(2)) geomDone = 0 gjfMulti_done = 1 elif geomDone != 1: tmp_m = self.pattern_blankLine.match(tmp_line) if tmp_m: lineEnd = lineNum tmp_geom = tmp_lines[lineStart+1: lineEnd] geomDone = 1 if geomDone != 1: print 'Sorry! The input file is not a standard .gjf file!' raise common_api.readGjfFileError else: pass # print 'Gjf file read in successfully!' gjfFile.close() self.mole = chem.molecule() try: self.mole.getGjfGeom(tmp_geom) except: print 'Sorry! The input file is not a standard .gjf file!' raise common_api.readGjfFileError self.mole.setSpinMultiplicity(tmp_multi) self.mole.calcFormula() tmp_m = self.pattern_formula.match(self.mole.formula) if tmp_m: if tmp_m.group(1) == '': carbonNumber = 1 else: carbonNumber = int(tmp_m.group(1)) if carbonNumber < 3: print 'Please submit a species with carbon atoms >= 3!' raise common_api.carbonLessThan3Error else: print 'Sorry! Not an alkane or alkene or their radical! Not supported temporarily!' raise common_api.beyondSpeciesRangeError if moleculeLabel == '': self.mole.setLabel(self.mole.formula) else: self.mole.setLabel(moleculeLabel) try: self.mole.fulfillBonds() except: print 'mole.fulfillBonds() failed!' raise common_api.readGjfFileError if self.mole.existRings(): print 'Sorry! Structure with rings is not supported temporarily!' raise common_api.ringExistingError return self.mole
tmp_m = pattern_D3Energy.match(tmp_line)
if tmp_m:
tmp_D3energy = float(tmp_m.group(1))
SPEnergy_done = 0
tmp_m = pattern_SPEnergy.match(tmp_line)
if tmp_m:
tmp_energy = float(tmp_m.group(1))
SPEnergy.append(tmp_energy - tmp_D3energy)
if np.abs(tmp_energy) > 1e-12 and SPEnergy_done == 0:
SPEnergy_done = 1
if SPEnergy_done != 1:
print 'Error! SP energy file error!' + tmp_file
energyFile.close()
tmp_mole = chem.molecule()
tmp_mole.getLogGeom(geom[-1])
tmp_mole.calcFormula()
formula.append(tmp_mole.formula)
atomsNum.append(tmp_mole.getAtomsNum())
name.append(tmp_name)
speciesNum += 1
###########################################
# write info to excel
###########################################
wb = xlsxwriter.Workbook('database.xlsx')
sh = wb.add_worksheet('speciesInfo')
tmp_row = 0
def setUp(self):
self.assy = assy = assembly.assembly(None)
self.mol = mol = chem.molecule(assy)
self.a = a = chem.Atom("C", chem.V(0.0, 0.0, 0.0), mol)
self.a.set_atomtype("sp3", True)
