def testRepeat(self):
self._readData()
nMols = len(self.smis)
for i in range(nMols):
smi = self.smis[i]
mol = Chem.MolFromSmiles(smi)
clog = self.clogs[i]
tmp = Crippen.MolLogP(mol)
tmp = Crippen.MolLogP(mol)
self.failUnless(feq(clog,tmp),'bad logp fooutF,r %s: %4.4f != %4.4f'%(smi,clog,tmp))
mr = self.mrs[i]
tmp = Crippen.MolMR(mol)
tmp = Crippen.MolMR(mol)
self.failUnless(feq(mr,tmp),'bad MR for %s: %4.4f != %4.4f'%(smi,mr,tmp))
def check_ligand(file_path):
bool = False
if os.path.isfile(file_path):
suppl = Chem.SDMolSupplier(file_path)
for mol in suppl:
if mol is not None:
# components of rule
hydrogen_bond_doner = True if Lipinski.NumHDonors(
mol) <= 5 else False
hydrogen_bond_acceptors = True if Lipinski.NumHAcceptors(
mol) <= 10 else False
molecular_mass = True if Descriptors.ExactMolWt(
mol) <= 500 else False
octanol_water_partition_coefficient_logP = True if Crippen.MolLogP(
mol) <= 5 else False
components_rank = hydrogen_bond_doner + hydrogen_bond_acceptors + molecular_mass + octanol_water_partition_coefficient_logP
# variants
partition_coefficient_logP = True if -0.4 <= Crippen.MolLogP(
mol) <= 5.6 else False
molar_refractivity = True if 40 <= Crippen.MolMR(
mol) <= 130 else False
molecular_weight = True if 180 <= Descriptors.ExactMolWt(
mol) <= 500 else False
number_of_atoms = True if 20 <= Lipinski.HeavyAtomCount(
mol) <= 70 else False
polar_surface_area = True if MolSurf.TPSA(
mol) <= 140 else False
variants_rank = partition_coefficient_logP + molar_refractivity + molecular_weight + number_of_atoms + polar_surface_area
if (components_rank == 4) and (variants_rank == 4
or variants_rank == 5):
bool = True
return bool
def testLogP(self):
self._readData()
nMols = len(self.smis)
#outF = file(self.fName,'w')
for i in range(nMols):
smi = self.smis[i]
mol = Chem.MolFromSmiles(smi)
if 1:
clog = self.clogs[i]
tmp = Crippen.MolLogP(mol)
self.failUnless(feq(clog,tmp),'bad logp for %s: %4.4f != %4.4f'%(smi,clog,tmp))
mr = self.mrs[i]
tmp = Crippen.MolMR(mol)
self.failUnless(feq(mr,tmp),'bad MR for %s: %4.4f != %4.4f'%(smi,mr,tmp))
else:
clog = Crippen.MolLogP(mol)
mr = Crippen.MolMR(mol)
print >>outF,'%s,%.4f,%.4f'%(smi,clog,mr)
def run_filter(self, mol):
"""
This runs a Ghose filter for drug-likeliness. Ghose filter filters
molecules by Molecular weight (MW), the number of atoms, and the logP
value.
We protonate the mol in this filter because hydrogens affect
atom count. Our Ghose implementation counts hydrogens in against
the total number of atoms.
To pass the filter a molecule must be:
MW between 160 and 480 dalton
Number of Atoms: between 20 and 70
logP between -0,4 and +5,6
Inputs:
:param rdkit.Chem.rdchem.Mol object mol: An rdkit mol object to be
tested if it passes the filters
Returns:
:returns: bool bool: True if the mol passes the filter; False if it
fails the filter
"""
# Make a copy of the mol so we can AddHs without affecting other filters
# number of atoms is altered by the presence/absence of hydrogens.
# Our Ghose filter counts hydrogenss towards atom count
copy_mol = copy.deepcopy(mol)
copy_mol = Chem.AddHs(copy_mol)
exact_mwt = Descriptors.ExactMolWt(copy_mol)
if ((exact_mwt < 160) or (exact_mwt > 480)):
return False
num_atoms = copy_mol.GetNumAtoms()
if ((num_atoms < 20) or (num_atoms > 70)):
return False
# molar Refractivity
MolMR = Crippen.MolMR(copy_mol)
if ((MolMR < 40) or (MolMR > 130)):
return False
# molar LogP
mol_log_p = Crippen.MolLogP(copy_mol)
if ((mol_log_p < -0.4) or (mol_log_p > 5.6)):
return False
# passed all filters
return True
def QSArproperties_test(array,forest, num, namefile):
##Bickerton, G.R.; Paolini, G.V.; Besnard, J.; Muresan, S.; Hopkins, A.L. (2012)
##Quantifying the chemical beauty of drugs,
##Nature Chemistry, 4, 90-98
##[https://doi.org/10.1038/nchem.1243]
#
##Wildman, S.A.; Crippen, G.M. (1999)
##Prediction of Physicochemical Parameters by Atomic Contributions,
##Journal of Chemical Information and Computer Sciences, 39, 868-873
##[https://doi.org/10.1021/ci990307l]
#
fw = open( 'QSAR-2D' + str(num) + str(namefile) + '.csv', 'w')
#
for line in array:
parameter= line.split(sep="\t",maxsplit=9)
peptide_seq = parameter[0]
peptide = parameter[1]
#
molpeptideLi = Chem.MolFromSmiles(peptide)
# subprocess
scoreSA_Li = calculateScore(molpeptideLi)
#
# Make Prediction Random-Forest
fp_vect_Li = genFP(molpeptideLi)
# Get probabilities
predictionsLi = forest.predict_proba(fp_vect_Li.reshape(1,-1))
#print ("Probability %s mutagenic %0.6f " % (sequence,predictionsLi[0][1]))
# See http://cdb.ics.uci.edu/cgibin/Smi2DepictWeb.py
propQSAR = QED.properties(molpeptideLi)
MolWeight = propQSAR.MW
MolLogP = propQSAR.ALOGP
HbondA = propQSAR.HBA
HbondD = propQSAR.HBD
PolarSA = propQSAR.PSA
Rbonds = propQSAR.ROTB
Aromatic = propQSAR.AROM
#
MolarRefractivity = Crippen.MolMR(molpeptideLi)
nAtoms = molpeptideLi.GetNumAtoms()
#
SynthAcces = scoreSA_Li
AmesMutagenic = predictionsLi[0][1]
#
result = ( str(MolWeight) + "\t" + str(MolLogP) + "\t" + str(HbondA) + "\t" + str(HbondD) + "\t" + str(PolarSA) + "\t" + str(Rbonds) + "\t" + str(MolarRefractivity) + "\t" + str(nAtoms) + "\t" + str(SynthAcces) + "\t" + str(AmesMutagenic) + "\t" + str (peptide_seq) + "\t" + str(peptide) + "\n")
#print (result)
fw.write(result)
fw.close()
def CalculateMolMR(mol):
"""
#################################################################
Cacluation of molecular refraction value based on Crippen method
---->MR
Usage:
result=CalculateMolMR(mol)
Input: mol is a molecule object.
Output: result is a numeric value.
#################################################################
"""
return round(Crippen.MolMR(mol), 3)
def PhyChem(smiles):
""" Calculating the 19D physicochemical descriptors for each molecules,
the value has been normalized with Gaussian distribution.
Arguments:
smiles (list): list of SMILES strings.
Returns:
props (ndarray): m X 19 matrix as normalized PhysChem descriptors.
m is the No. of samples
"""
props = []
for smile in smiles:
mol = Chem.MolFromSmiles(smile)
try:
MW = desc.MolWt(mol)
LOGP = Crippen.MolLogP(mol)
HBA = Lipinski.NumHAcceptors(mol)
HBD = Lipinski.NumHDonors(mol)
rotable = Lipinski.NumRotatableBonds(mol)
amide = AllChem.CalcNumAmideBonds(mol)
bridge = AllChem.CalcNumBridgeheadAtoms(mol)
heteroA = Lipinski.NumHeteroatoms(mol)
heavy = Lipinski.HeavyAtomCount(mol)
spiro = AllChem.CalcNumSpiroAtoms(mol)
FCSP3 = AllChem.CalcFractionCSP3(mol)
ring = Lipinski.RingCount(mol)
Aliphatic = AllChem.CalcNumAliphaticRings(mol)
aromatic = AllChem.CalcNumAromaticRings(mol)
saturated = AllChem.CalcNumSaturatedRings(mol)
heteroR = AllChem.CalcNumHeterocycles(mol)
TPSA = MolSurf.TPSA(mol)
valence = desc.NumValenceElectrons(mol)
mr = Crippen.MolMR(mol)
# charge = AllChem.ComputeGasteigerCharges(mol)
prop = [
MW, LOGP, HBA, HBD, rotable, amide, bridge, heteroA, heavy,
spiro, FCSP3, ring, Aliphatic, aromatic, saturated, heteroR,
TPSA, valence, mr
]
except Exception:
print(smile)
prop = [0] * 19
props.append(prop)
props = np.array(props)
props = Scaler().fit_transform(props)
return props
def QSArproperties(mol, sa, seq, Ames):
propQSAR = QED.properties(mol)
"""Bickerton, G.R.; Paolini, G.V.; Besnard, J.; Muresan, S.; Hopkins, A.L. (2012)
Quantifying the chemical beauty of drugs,
Nature Chemistry, 4, 90-98
[https://doi.org/10.1038/nchem.1243]"""
#
"""Wildman, S.A.; Crippen, G.M. (1999)
Prediction of Physicochemical Parameters by Atomic Contributions,
Journal of Chemical Information and Computer Sciences, 39, 868-873
[https://doi.org/10.1021/ci990307l]"""
#
MolWeight = propQSAR.MW
MolLogP = propQSAR.ALOGP
HbondA = propQSAR.HBA
HbondD = propQSAR.HBD
PolarSA = propQSAR.PSA
Rbonds = propQSAR.ROTB
Aromatic = propQSAR.AROM
MolarRefractivity = Crippen.MolMR(mol)
SynthAcces = sa
nAtoms = mol.GetNumAtoms()
AmesMutagenic = Ames
def extract(x, from_smiles):
if from_smiles:
mol = Chem.MolFromSmiles(x)
else:
mol = x
if (mol is None) or (len(mol.GetAtoms()) == 0):
if include_3D:
return [0] * 29
else:
return [0] * 24
else:
logP = Crippen.MolLogP(mol)
refractivity = Crippen.MolMR(mol)
weight = Descriptors.MolWt(mol)
exact_weight = Descriptors.ExactMolWt(mol)
heavy_weight = Descriptors.HeavyAtomMolWt(mol)
heavy_count = Lipinski.HeavyAtomCount(mol)
nhoh_count = Lipinski.NHOHCount(mol)
no_count = Lipinski.NOCount(mol)
hacceptor_count = Lipinski.NumHAcceptors(mol)
hdonor_count = Lipinski.NumHDonors(mol)
hetero_count = Lipinski.NumHeteroatoms(mol)
rotatable_bond_count = Lipinski.NumRotatableBonds(mol)
valance_electron_count = Descriptors.NumValenceElectrons(mol)
amide_bond_count = rdMolDescriptors.CalcNumAmideBonds(mol)
aliphatic_ring_count = Lipinski.NumAliphaticRings(mol)
aromatic_ring_count = Lipinski.NumAromaticRings(mol)
saturated_ring_count = Lipinski.NumSaturatedRings(mol)
aliphatic_cycle_count = Lipinski.NumAliphaticCarbocycles(mol)
aliphaticHetero_cycle_count = Lipinski.NumAliphaticHeterocycles(
mol)
aromatic_cycle_count = Lipinski.NumAromaticCarbocycles(mol)
aromaticHetero_cycle_count = Lipinski.NumAromaticHeterocycles(mol)
saturated_cycle_count = Lipinski.NumSaturatedCarbocycles(mol)
saturatedHetero_cycle_count = Lipinski.NumSaturatedHeterocycles(
mol)
tpsa = rdMolDescriptors.CalcTPSA(mol)
if include_3D:
mol_3D = Chem.AddHs(mol)
AllChem.EmbedMolecule(mol_3D)
AllChem.MMFFOptimizeMolecule(mol_3D)
eccentricity = rdMolDescriptors.CalcEccentricity(mol_3D)
asphericity = rdMolDescriptors.CalcAsphericity(mol_3D)
spherocity = rdMolDescriptors.CalcSpherocityIndex(mol_3D)
inertial = rdMolDescriptors.CalcInertialShapeFactor(mol_3D)
gyration = rdMolDescriptors.CalcRadiusOfGyration(mol_3D)
return [
logP, refractivity, weight, exact_weight, heavy_weight,
heavy_count, nhoh_count, no_count, hacceptor_count,
hdonor_count, hetero_count, rotatable_bond_count,
valance_electron_count, amide_bond_count,
aliphatic_ring_count, aromatic_ring_count,
saturated_ring_count, aliphatic_cycle_count,
aliphaticHetero_cycle_count, aromatic_cycle_count,
aromaticHetero_cycle_count, saturated_cycle_count,
saturatedHetero_cycle_count, tpsa, eccentricity,
asphericity, spherocity, inertial, gyration
]
else:
return [
logP, refractivity, weight, exact_weight, heavy_weight,
heavy_count, nhoh_count, no_count, hacceptor_count,
hdonor_count, hetero_count, rotatable_bond_count,
valance_electron_count, amide_bond_count,
aliphatic_ring_count, aromatic_ring_count,
saturated_ring_count, aliphatic_cycle_count,
aliphaticHetero_cycle_count, aromatic_cycle_count,
aromaticHetero_cycle_count, saturated_cycle_count,
saturatedHetero_cycle_count, tpsa
]
def mr(mol):
return Crippen.MolMR(mol) if mol else None
def calc_rdkit(mol):
descriptors = pd.Series(
np.array([
Crippen.MolLogP(mol),
Crippen.MolMR(mol),
Descriptors.FpDensityMorgan1(mol),
Descriptors.FpDensityMorgan2(mol),
Descriptors.FpDensityMorgan3(mol),
Descriptors.FractionCSP3(mol),
Descriptors.HeavyAtomMolWt(mol),
Descriptors.MaxAbsPartialCharge(mol),
Descriptors.MaxPartialCharge(mol),
Descriptors.MinAbsPartialCharge(mol),
Descriptors.MinPartialCharge(mol),
Descriptors.MolWt(mol),
Descriptors.NumRadicalElectrons(mol),
Descriptors.NumValenceElectrons(mol),
EState.EState.MaxAbsEStateIndex(mol),
EState.EState.MaxEStateIndex(mol),
EState.EState.MinAbsEStateIndex(mol),
EState.EState.MinEStateIndex(mol),
EState.EState_VSA.EState_VSA1(mol),
EState.EState_VSA.EState_VSA10(mol),
EState.EState_VSA.EState_VSA11(mol),
EState.EState_VSA.EState_VSA2(mol),
EState.EState_VSA.EState_VSA3(mol),
EState.EState_VSA.EState_VSA4(mol),
EState.EState_VSA.EState_VSA5(mol),
EState.EState_VSA.EState_VSA6(mol),
EState.EState_VSA.EState_VSA7(mol),
EState.EState_VSA.EState_VSA8(mol),
EState.EState_VSA.EState_VSA9(mol),
Fragments.fr_Al_COO(mol),
Fragments.fr_Al_OH(mol),
Fragments.fr_Al_OH_noTert(mol),
Fragments.fr_aldehyde(mol),
Fragments.fr_alkyl_carbamate(mol),
Fragments.fr_alkyl_halide(mol),
Fragments.fr_allylic_oxid(mol),
Fragments.fr_amide(mol),
Fragments.fr_amidine(mol),
Fragments.fr_aniline(mol),
Fragments.fr_Ar_COO(mol),
Fragments.fr_Ar_N(mol),
Fragments.fr_Ar_NH(mol),
Fragments.fr_Ar_OH(mol),
Fragments.fr_ArN(mol),
Fragments.fr_aryl_methyl(mol),
Fragments.fr_azide(mol),
Fragments.fr_azo(mol),
Fragments.fr_barbitur(mol),
Fragments.fr_benzene(mol),
Fragments.fr_benzodiazepine(mol),
Fragments.fr_bicyclic(mol),
Fragments.fr_C_O(mol),
Fragments.fr_C_O_noCOO(mol),
Fragments.fr_C_S(mol),
Fragments.fr_COO(mol),
Fragments.fr_COO2(mol),
Fragments.fr_diazo(mol),
Fragments.fr_dihydropyridine(mol),
Fragments.fr_epoxide(mol),
Fragments.fr_ester(mol),
Fragments.fr_ether(mol),
Fragments.fr_furan(mol),
Fragments.fr_guanido(mol),
Fragments.fr_halogen(mol),
Fragments.fr_hdrzine(mol),
Fragments.fr_hdrzone(mol),
Fragments.fr_HOCCN(mol),
Fragments.fr_imidazole(mol),
Fragments.fr_imide(mol),
Fragments.fr_Imine(mol),
Fragments.fr_isocyan(mol),
Fragments.fr_isothiocyan(mol),
Fragments.fr_ketone(mol),
Fragments.fr_ketone_Topliss(mol),
Fragments.fr_lactam(mol),
Fragments.fr_lactone(mol),
Fragments.fr_methoxy(mol),
Fragments.fr_morpholine(mol),
Fragments.fr_N_O(mol),
Fragments.fr_Ndealkylation1(mol),
Fragments.fr_Ndealkylation2(mol),
Fragments.fr_NH0(mol),
Fragments.fr_NH1(mol),
Fragments.fr_NH2(mol),
Fragments.fr_Nhpyrrole(mol),
Fragments.fr_nitrile(mol),
Fragments.fr_nitro(mol),
Fragments.fr_nitro_arom(mol),
Fragments.fr_nitro_arom_nonortho(mol),
Fragments.fr_nitroso(mol),
Fragments.fr_oxazole(mol),
Fragments.fr_oxime(mol),
Fragments.fr_para_hydroxylation(mol),
Fragments.fr_phenol(mol),
Fragments.fr_phenol_noOrthoHbond(mol),
Fragments.fr_phos_acid(mol),
Fragments.fr_phos_ester(mol),
Fragments.fr_piperdine(mol),
Fragments.fr_piperzine(mol),
Fragments.fr_priamide(mol),
Fragments.fr_prisulfonamd(mol),
Fragments.fr_pyridine(mol),
Fragments.fr_quatN(mol),
Fragments.fr_SH(mol),
Fragments.fr_sulfide(mol),
Fragments.fr_sulfonamd(mol),
Fragments.fr_sulfone(mol),
Fragments.fr_term_acetylene(mol),
Fragments.fr_tetrazole(mol),
Fragments.fr_thiazole(mol),
Fragments.fr_thiocyan(mol),
Fragments.fr_thiophene(mol),
Fragments.fr_unbrch_alkane(mol),
Fragments.fr_urea(mol),
GraphDescriptors.BalabanJ(mol),
GraphDescriptors.BertzCT(mol),
GraphDescriptors.Chi0(mol),
GraphDescriptors.Chi0n(mol),
GraphDescriptors.Chi0v(mol),
GraphDescriptors.Chi1(mol),
GraphDescriptors.Chi1n(mol),
GraphDescriptors.Chi1v(mol),
GraphDescriptors.Chi2n(mol),
GraphDescriptors.Chi2v(mol),
GraphDescriptors.Chi3n(mol),
GraphDescriptors.Chi3v(mol),
GraphDescriptors.Chi4n(mol),
GraphDescriptors.Chi4v(mol),
GraphDescriptors.HallKierAlpha(mol),
GraphDescriptors.Ipc(mol),
GraphDescriptors.Kappa1(mol),
GraphDescriptors.Kappa2(mol),
GraphDescriptors.Kappa3(mol),
Lipinski.HeavyAtomCount(mol),
Lipinski.NHOHCount(mol),
Lipinski.NOCount(mol),
Lipinski.NumAliphaticCarbocycles(mol),
Lipinski.NumAliphaticHeterocycles(mol),
Lipinski.NumAliphaticRings(mol),
Lipinski.NumAromaticCarbocycles(mol),
Lipinski.NumAromaticHeterocycles(mol),
Lipinski.NumAromaticRings(mol),
Lipinski.NumHAcceptors(mol),
Lipinski.NumHDonors(mol),
Lipinski.NumHeteroatoms(mol),
Lipinski.NumRotatableBonds(mol),
Lipinski.NumSaturatedCarbocycles(mol),
Lipinski.NumSaturatedHeterocycles(mol),
Lipinski.NumSaturatedRings(mol),
Lipinski.RingCount(mol),
MolSurf.LabuteASA(mol),
MolSurf.PEOE_VSA1(mol),
MolSurf.PEOE_VSA10(mol),
MolSurf.PEOE_VSA11(mol),
MolSurf.PEOE_VSA12(mol),
MolSurf.PEOE_VSA13(mol),
MolSurf.PEOE_VSA14(mol),
MolSurf.PEOE_VSA2(mol),
MolSurf.PEOE_VSA3(mol),
MolSurf.PEOE_VSA4(mol),
MolSurf.PEOE_VSA5(mol),
MolSurf.PEOE_VSA6(mol),
MolSurf.PEOE_VSA7(mol),
MolSurf.PEOE_VSA8(mol),
MolSurf.PEOE_VSA9(mol),
MolSurf.SlogP_VSA1(mol),
MolSurf.SlogP_VSA10(mol),
MolSurf.SlogP_VSA11(mol),
MolSurf.SlogP_VSA12(mol),
MolSurf.SlogP_VSA2(mol),
MolSurf.SlogP_VSA3(mol),
MolSurf.SlogP_VSA4(mol),
MolSurf.SlogP_VSA5(mol),
MolSurf.SlogP_VSA6(mol),
MolSurf.SlogP_VSA7(mol),
MolSurf.SlogP_VSA8(mol),
MolSurf.SlogP_VSA9(mol),
MolSurf.SMR_VSA1(mol),
MolSurf.SMR_VSA10(mol),
MolSurf.SMR_VSA2(mol),
MolSurf.SMR_VSA3(mol),
MolSurf.SMR_VSA4(mol),
MolSurf.SMR_VSA5(mol),
MolSurf.SMR_VSA6(mol),
MolSurf.SMR_VSA7(mol),
MolSurf.SMR_VSA8(mol),
MolSurf.SMR_VSA9(mol),
MolSurf.TPSA(mol)
]))
return descriptors
def calculate(self):
return Crippen.MolMR(self.mol)
def CalculateMolMR(mol):
return round(Crippen.MolMR(mol), 3)
