def ExecuteOperation(vStack,opStack, subformulas):
op = opStack.pop()
rightOperand = vStack.pop()
if(op.valor==u'￢'):
subFormula = Tokenizer(''+ op.valor +rightOperand.valor, "Subformula" )
subFormula.setValue(not(rightOperand.booleano))
vStack.append(subFormula)
subformulas.append(subFormula)
return vStack, opStack, subformulas
leftOperand = vStack.pop()
subFormula = Tokenizer(''+leftOperand.valor + op.valor + rightOperand.valor, "Subformula" )
if(op.valor==u'->'):
if(leftOperand.booleano==False or rightOperand.booleano == True):
subFormula.setValue(True)
else:
subFormula.setValue(False)
elif(op.valor==u'<->'):
if(leftOperand.booleano==rightOperand.booleano):
subFormula.setValue(True)
else:
subFormula.setValue(False)
elif(op.valor==u'∨'):
subFormula.setValue(leftOperand.booleano or rightOperand.booleano)
elif(op.valor==u'∧'):
subFormula.setValue(leftOperand.booleano and rightOperand.booleano)
vStack.append(subFormula)
subformulas.append(subFormula)
return vStack, opStack, subformulas
def retiraToken(expressao,i):
if(expressao[i]==" " or expressao[i]=="\n"):
i = i + 1;
if(expressao[i]==u'∨' or expressao[i]==u'∧'):
Token = Tokenizer(expressao[i],"Conectivos")
return Token, i+1
elif(expressao[i]=="t"):
if(expressao[i:i+4]=="true"):
Token = Tokenizer(expressao[i:i+4],"BooleanSymbols")
Token.setValue(True)
return Token, i+4
else:
Token = Tokenizer(expressao[i],"Proposicionais")
return Token, i + 1
elif(expressao[i]=="f"):
if(expressao[i:i+5]=="false"):
Token = Tokenizer(expressao[i:i+5],"BooleanSymbols")
Token.setValue(False)
return Token, i+5
else:
Token = Tokenizer(expressao[i],"Proposicionais")
return Token, i + 1
elif(expressao[i] =="(" or expressao[i]==")"):
Token = Tokenizer(expressao[i],"Pontuacao")
return Token, i+1
elif(expressao[i]==u"￢"):
Token = Tokenizer(expressao[i],"Negacao")
return Token, i+1
elif(expressao[i].isalpha()):
Token = Tokenizer(expressao[i],"Proposicionais")
return Token, i + 1
elif(expressao[i]==u'<'):
if(expressao[i:i+3]==u'<->'):
Token = Tokenizer(expressao[i:i+3],"Conectivos")
return Token, i+3
else:
sys.exit("Cadeia Inválida")
elif(expressao[i]==u'-'):
if(expressao[i:i+2]==u'->'):
Token = Tokenizer(expressao[i:i+2],"Conectivos")
return Token, i+2
else:
sys.exit("Cadeia Inválida")
else:
sys.exit("Cadeia Inválida")
mol = self.nrm.normalize(mol)
mol = self.lfc.choose(mol)
mol = self.uc.uncharge(mol)
return Chem.MolToSmiles(mol, isomericSmiles=False, canonical=True)
else:
return None
if __name__ == "__main__":
with open('data/canonical_smiles.smi', 'r') as file:
smiles = [line.rstrip() for line in file]
print("Initial number of sequences %i" % len(smiles))
p = Preprocess()
t = Tokenizer()
# Normalization, uncharging, removing chirality and light fragments
nn_smi = [p.clean(smile) for smile in tqdm(smiles)]
unn_smi = list(set([smile for smile in nn_smi if smile]))
# Limit sequence length 34-128
cl_smi = []
for smile in unn_smi:
if 34 <= len(t.tokenize(smile)) <= 128:
cl_smi.append(smile)
print("Number of sequences after cleaning %i" % len(cl_smi))
with open('data/cleaned_smiles.smi', 'w') as file:
for line in cl_smi:
