def extract_named_descriptors_of_seq(sequence):
'''
Returns a map ("descriptor" -> value) of descriptors when given a sequence of aminoacids (string)
:param sequence:
:return:
'''
#mol = Chem.MolFromSequence(str(sequence))
res = {}
sequence = str(sequence)
if len(sequence) > 3:
#print(Autocorrelation.CalculateGearyAutoTotal(sequence))
res = PseudoAAC.GetPseudoAAC(sequence,
lamda=3,
weight=0.05,
AAP=[
_Hydrophobicity, _hydrophilicity,
_residuemass, _pK1, _pK2, _pI
])
res.update(AAComposition_extra_PS.CalculateAAComposition(sequence))
res.update(CTD.CalculateCTD(sequence))
mol = Chem.MolFromSequence(str(sequence))
#res = geary.GetGearyAuto(mol)
res.update(kappa.GetKappa(mol))
res.update(charge.GetCharge(mol))
#res.update(moran.GetMoranAuto(mol))
res.update(moreaubroto.GetMoreauBrotoAuto(mol))
res.update(molproperty.GetMolecularProperty(mol))
#res.update(moe.GetMOE(mol))
res.update(basak.Getbasak(mol))
#print(res)
#input()
return res
def main_process(self, str_pep_file, str_des_out):
self.str_pep_file = str_pep_file
self.str_des_out = str_des_out
my_pep, list_class_label = Str_DS_class().read_pep_file(
self.str_pep_file)
Str_DS_class().structure_gen(my_pep)
sValues = []
sdf_names = glob.glob(os.path.join("./strs", '*.sdf'))
n = len(sdf_names)
#print sdf_names
print "Descriptors being calculated !"
bar2 = Bar('Processing SDF file', fill='>', max=n)
for sdn in my_pep:
mols = Chem.SDMolSupplier(os.path.join("./strs", sdn + ".sdf"))
#mols = Chem.SDMolSupplier(sdn)
for mol in mols:
sValue = Str_DS_class().new_Des_gene(mol)
sValues.append(sValue)
bar2.next()
sDF1 = pd.DataFrame(sValues)
#sDF1.to_csv(self.str_des_out, index = False)
bar2.finish()
print "Descripor calculation has been finished "
return sDF1, list_class_label
def smiles_converter(self, smi, L):
res = []
try:
mol = Chem.MolFromSmiles(smi)
res.append(constitution.GetConstitutional(mol))
res.append(connectivity.GetConnectivity(mol))
res.append(fingerprint.CalculateMACCSFingerprint(mol)[1])
try:
res.append(kappa.GetKappa(mol))
except:
pass
try:
drug.ReadMolFromSmile(smi)
res.append(drug.GetMOE())
except:
pass
try:
res.append(charge.GetCharge(mol))
except:
pass
except:
pass
super_dict = {} # uses set to avoid duplicates
if len(res) >= 1:
for d in res:
for k, v in d.items(): # use d.iteritems() in python 2
super_dict[k] = v
super_dict["smiles"] = smi.strip()
L.append(super_dict)
def get_f1_f2(row):
try:
protein_sequence = dpi.GetProteinSequenceFromID(row['target_id'])
dpi.ReadProteinSequence(protein_sequence)
aa_composition = dpi.GetAAComp() #COMPOSICAO AMINOACIOS
molecule = Chem.MolFromSmiles(row['smiles'])
kappa_descriptors = kappa.GetKappa(molecule)
if (row.name % 500 == 0
): # para facilitar o processo, a leitura e feita aos poucos
partial = pd.DataFrame(
result_list_F1
) # os smiles e target_id das colunas que dao erros que sao guardados num ficheiro
partial.to_csv(export_path + "export_partial_f1.csv")
partial = pd.DataFrame(result_list_F2)
partial.to_csv(export_path + "export_partial_f2.csv")
partial = pd.DataFrame(target_id_errors)
partial.to_csv(export_path + "errors.csv")
result_list_F1.append(
dpi.GetDPIFeature1(kappa_descriptors, aa_composition))
result_list_F2.append(
dpi.GetDPIFeature2(kappa_descriptors, aa_composition))
except:
dic = {'smiles': row['smiles'], 'target_id': row['target_id']}
target_id_errors.append(dic)
print(row.name)