def lipinski_rule(mol):
fingerprint = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(mol)
return [
Lipinski.NHOHCount(mol) <= 5,
Lipinski.NOCount(mol) <= 10,
Descriptors.ExactMolWt(mol) <= 500,
LogP('logP').run(fingerprint) <= 5]
def GenerateAtomPairsFingerprints(Mols):
"""Generate AtomPairs fingerprints."""
MiscUtil.PrintInfo("\nGenerating AtomPairs fingerprints...")
MinLength = OptionsInfo["FingerprintsParams"]["AtomPairs"]["MinLength"]
MaxLength = OptionsInfo["FingerprintsParams"]["AtomPairs"]["MaxLength"]
UseChirality = OptionsInfo["FingerprintsParams"]["AtomPairs"][
"UseChirality"]
if OptionsInfo["GenerateBitVectFingerints"]:
# Generate ExplicitBitVect fingerprints...
FPSize = 2048
BitsPerHash = 4
MolsFingerprints = [
rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
Mol,
minLength=MinLength,
maxLength=MaxLength,
includeChirality=UseChirality,
nBits=FPSize,
nBitsPerEntry=BitsPerHash) for Mol in Mols
]
else:
# Generate IntSparseIntVect fingerprints...
MolsFingerprints = [
rdMolDescriptors.GetAtomPairFingerprint(
Mol,
minLength=MinLength,
maxLength=MaxLength,
includeChirality=UseChirality) for Mol in Mols
]
return MolsFingerprints
def __init__(self, fp_type, fp_bits=2048):
"""
:param fp_type: fingerprint type
:param fp_bits: number of fingerprint bits
"""
self.fp_type = fp_type
self.fp_dict = {}
self.fp_dict['morgan2'] = [
lambda m: rdmd.GetMorganFingerprintAsBitVect(m, 2, nBits=fp_bits),
fp_bits
]
self.fp_dict['morgan3'] = [
lambda m: rdmd.GetMorganFingerprintAsBitVect(m, 3, nBits=fp_bits),
fp_bits
]
self.fp_dict['ap'] = [
lambda m: rdmd.GetHashedAtomPairFingerprintAsBitVect(
m, nBits=fp_bits), fp_bits
]
self.fp_dict['rdk5'] = [
lambda m: Chem.RDKFingerprint(
m, maxPath=5, fpSize=fp_bits, nBitsPerHash=2), fp_bits
]
if self.fp_dict.get(fp_type):
self.fp_function = self.fp_dict[fp_type]
else:
print("invalid fingerprint type: %s" % fp_type)
sys.exit(0)
def pka_similarities(smile, mol_set, n):
mol = Chem.MolFromSmiles(smile)
mol_fp = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(mol)
similarity = []
for molecule in mol_set:
sim = DataStructs.DiceSimilarity(mol_fp, molecule[2])
similarity.append([sim, molecule[1]])
return np.asarray(sorted(similarity)[:n]).flatten()
def get_atompairs(molecule, length=512):
try:
atompairs = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
molecule, nBits=length)
except Exception as e:
print(e)
print("error" + str(molecule))
atompairs = np.nan
return atompairs
def testHashedAtomPairs(self):
m = Chem.MolFromSmiles('c1ccccc1')
fp1 = rdMD.GetHashedAtomPairFingerprint(m, 2048)
fp2 = rdMD.GetHashedAtomPairFingerprint(m, 2048, 1, 3)
self.assertTrue(fp1 == fp2)
fp2 = rdMD.GetHashedAtomPairFingerprint(m, 2048, 1, 2)
sim = DataStructs.DiceSimilarity(fp1, fp2)
self.assertTrue(sim > 0.0 and sim < 1.0)
m = Chem.MolFromSmiles('c1ccccn1')
fp2 = rdMD.GetHashedAtomPairFingerprint(m, 2048)
sim = DataStructs.DiceSimilarity(fp1, fp2)
self.assertTrue(sim > 0.0 and sim < 1.0)
m = Chem.MolFromSmiles('c1ccccc1')
fp1 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m, 2048)
m = Chem.MolFromSmiles('c1ccccn1')
fp2 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m, 2048)
sim = DataStructs.DiceSimilarity(fp1, fp2)
self.assertTrue(sim > 0.0 and sim < 1.0)
def featurize(self, x):
if self.input_type == 'smiles':
x_ = x
x = Chem.MolFromSmiles(x)
if x is None:
raise ValueError('can not convert Mol from SMILES %s' % x_)
if self.input_type == 'any':
if not isinstance(x, Chem.rdchem.Mol):
x_ = x
x = Chem.MolFromSmiles(x)
if x is None:
raise ValueError('can not convert Mol from SMILES %s' % x_)
return list(
rdMol.GetHashedAtomPairFingerprintAsBitVect(x, nBits=self.n_bits))
def _encode(smi: str, fingerprint: str, radius: int,
length: int) -> T_comp:
"""fingerprint functions must be wrapped in a static function
so that they may be pickled for parallel processing
Parameters
----------
smi : str
the SMILES string of the molecule to encode
fingerprint : str
the the type of fingerprint to generate
radius : int
the radius of the fingerprint
length : int
the length of the fingerprint
Returns
-------
T_comp
the compressed feature representation of the molecule
"""
mol = Chem.MolFromSmiles(smi)
if fingerprint == 'morgan':
return rdmd.GetMorganFingerprintAsBitVect(mol,
radius=radius,
nBits=length,
useChirality=True)
if fingerprint == 'pair':
return rdmd.GetHashedAtomPairFingerprintAsBitVect(mol,
minLength=1,
maxLength=1 +
radius,
nBits=length)
if fingerprint == 'rdkit':
return rdmd.RDKFingerprint(mol,
minPath=1,
maxPath=1 + radius,
fpSize=length)
if fingerprint == 'maccs':
return rdmd.GetMACCSKeysFingerprint(mol)
if fingerprint == 'map4':
return map4.MAP4Calculator(dimensions=length,
radius=radius,
is_folded=True).calculate(mol)
raise NotImplementedError(f'Unrecognized fingerprint: "{fingerprint}"')
def predict(self, mol, selected_descriptors):
options = [0, 0, 0, 0, 0]
return_properties = {}
for option in selected_descriptors:
if option == 'logP':
options[0] = 1
elif option == 'sol':
options[0] = 1
options[1] = 1
elif option == 'mp':
options[0] = 1
options[1] = 1
options[2] = 1
elif option == 'pka':
options[3] = 1
elif option == 'mol_wt':
options[4] = 1
fp = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(mol)
if options[0]:
logP = self.logP_model.run(fp)
return_properties['logP'] = logP
if options[1]:
logP_sol = self.logP_solubility_model.run(logP)
atom_pair_sol = self.atom_pair_sol_model.run(fp)
combined_sol = self.combined_model.run(mol, logP,
logP_sol, atom_pair_sol)
mg_ml_sol = logs_to_mg_ml(combined_sol, mol)
return_properties['sol'] = mg_ml_sol
if options[2]:
mp = self.melting_point_model.run(combined_sol, logP)
return_properties['mp'] = mp
if options[3]:
avalon = GetAvalonFP(mol)
maacs = MACCSkeys.GenMACCSKeys(mol)
pka = self.pKa_model.run(avalon + maacs + fp)
return_properties['pka'] = pka
if options[4]:
wt = rdMolDescriptors.CalcExactMolWt(mol)
return_properties['mol_wt'] = wt
return return_properties
def featurize(self, x):
if self.input_type == 'smiles':
x_ = x
x = Chem.MolFromSmiles(x)
if x is None:
raise ValueError('cannot convert Mol from SMILES %s' % x_)
if self.input_type == 'any':
if not isinstance(x, Chem.rdchem.Mol):
x_ = x
x = Chem.MolFromSmiles(x)
if x is None:
raise ValueError('cannot convert Mol from SMILES %s' % x_)
if self.counting:
return count_fp(rdMol.GetHashedAtomPairFingerprint(x, nBits=self.n_bits), dim=self.n_bits)
else:
return list(rdMol.GetHashedAtomPairFingerprintAsBitVect(x, nBits=self.n_bits,
nBitsPerEntry=self.bit_per_entry))
def testAtomPairOptions(self):
m1 = Chem.MolFromSmiles('c1ccccc1')
m2 = Chem.MolFromSmiles('c1ccccn1')
fp1 = rdMD.GetAtomPairFingerprint(m1)
fp2 = rdMD.GetAtomPairFingerprint(m2)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetAtomPairFingerprint(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetAtomPairFingerprint(m2,atomInvariants=[1]*6)
self.assertEqual(fp1,fp2)
fp1 = rdMD.GetAtomPairFingerprint(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetAtomPairFingerprint(m2,atomInvariants=[2]*6)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m1)
fp2 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m2)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m2,atomInvariants=[1]*6)
self.assertEqual(fp1,fp2)
fp1 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetHashedAtomPairFingerprintAsBitVect(m2,atomInvariants=[2]*6)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetTopologicalTorsionFingerprint(m1)
fp2 = rdMD.GetTopologicalTorsionFingerprint(m2)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetTopologicalTorsionFingerprint(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetTopologicalTorsionFingerprint(m2,atomInvariants=[1]*6)
self.assertEqual(fp1,fp2)
fp1 = rdMD.GetTopologicalTorsionFingerprint(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetTopologicalTorsionFingerprint(m2,atomInvariants=[2]*6)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetHashedTopologicalTorsionFingerprintAsBitVect(m1)
fp2 = rdMD.GetHashedTopologicalTorsionFingerprintAsBitVect(m2)
self.assertNotEqual(fp1,fp2)
fp1 = rdMD.GetHashedTopologicalTorsionFingerprintAsBitVect(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetHashedTopologicalTorsionFingerprintAsBitVect(m2,atomInvariants=[1]*6)
self.assertEqual(fp1,fp2)
fp1 = rdMD.GetHashedTopologicalTorsionFingerprintAsBitVect(m1,atomInvariants=[1]*6)
fp2 = rdMD.GetHashedTopologicalTorsionFingerprintAsBitVect(m2,atomInvariants=[2]*6)
self.assertNotEqual(fp1,fp2)
def GenerateAtomPairsFingerprints(Mols):
"""Generate AtomPairs fingerprints."""
MiscUtil.PrintInfo("\nGenerating AtomPairs %s fingerprints..." % OptionsInfo["SpecifiedFingerprintsType"])
MinLength = OptionsInfo["FingerprintsParams"]["AtomPairs"]["MinLength"]
MaxLength = OptionsInfo["FingerprintsParams"]["AtomPairs"]["MaxLength"]
UseChirality = OptionsInfo["FingerprintsParams"]["AtomPairs"]["UseChirality"]
FPSize = OptionsInfo["FingerprintsParams"]["AtomPairs"]["FPSize"]
BitsPerHash = OptionsInfo["FingerprintsParams"]["AtomPairs"]["BitsPerHash"]
if re.match("^BitVect$", OptionsInfo["SpecifiedFingerprintsType"], re.I):
# Generate ExplicitBitVect fingerprints...
MiscUtil.PrintInfo("FPSize: %s; BitsPerHash: %s" % (FPSize, BitsPerHash))
MolsFingerprints = [rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(Mol, minLength = MinLength, maxLength = MaxLength, includeChirality = UseChirality, nBits = FPSize, nBitsPerEntry = BitsPerHash) for Mol in Mols]
else:
# Generate IntSparseIntVect fingerprints...
MolsFingerprints = [rdMolDescriptors.GetAtomPairFingerprint(Mol, minLength = MinLength, maxLength = MaxLength, includeChirality = UseChirality) for Mol in Mols]
return MolsFingerprints
def __init__(self, fp_type_list, num_fp_bits=1024):
self.num_fp_bits = num_fp_bits
self.fp_function_list = []
self.fp_type_list = fp_type_list
self.fp_dict = {}
self.des_names = [name[0] for name in Descriptors._descList]
des_calculator = MoleculeDescriptors.MolecularDescriptorCalculator(
self.des_names)
self.fp_dict['descriptors'] = [
lambda m: des_calculator.CalcDescriptors(m), -1
]
self.fp_dict['morgan2'] = [
lambda m: rdmd.GetMorganFingerprintAsBitVect(
m, 2, nBits=self.num_fp_bits), self.num_fp_bits
]
self.fp_dict['morgan3'] = [
lambda m: rdmd.GetMorganFingerprintAsBitVect(
m, 3, nBits=self.num_fp_bits), self.num_fp_bits
]
self.fp_dict['ap'] = [
lambda m: rdmd.GetHashedAtomPairFingerprintAsBitVect(
m, nBits=self.num_fp_bits), self.num_fp_bits
]
self.fp_dict['rdk5'] = [
lambda m: Chem.RDKFingerprint(
m, maxPath=5, fpSize=self.num_fp_bits, nBitsPerHash=2),
self.num_fp_bits
]
self.fp_names = []
for fp_type in fp_type_list:
if self.fp_dict.get(fp_type):
self.fp_function_list.append(self.fp_dict[fp_type])
if fp_type == "descriptors":
self.fp_names += self.des_names
else:
self.fp_names += self.get_names(self.num_fp_bits)
else:
print("invalid fingerprint type: %s" % fp_type)
sys.exit(1)
from chemical_models import AtomPairSolubility, LogP, LogPSolubility
data = open('data/water_solubility/aqsol.txt', 'r')
logP_model = LogP('logP')
logP_solubility_model = LogPSolubility('logS_logP')
atom_pair_sol_model = AtomPairSolubility('water_solubility')
X = []
Y = []
for line in data.readlines():
split = line.split(' ')
mol = Chem.MolFromSmiles(split[0])
fingerprint = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(mol)
logP = logP_model.run(fingerprint)
logP_sol = logP_solubility_model.run(logP)
atom_pair_sol = atom_pair_sol_model.run(fingerprint)
# Additional ESOL empirical model to increase accuracy
mw = Descriptors.ExactMolWt(mol)
rb = rdMolDescriptors.CalcNumRotatableBonds(mol)
ap = len(mol.GetSubstructMatches(
Chem.MolFromSmarts('[a]'))) / mol.GetNumHeavyAtoms()
esol = 0.16 - 0.63 * logP - 0.0062 * mw + 0.066 * rb - 0.74 * ap
X.append([logP_sol, atom_pair_sol, esol])
Y.append(float(split[1][:-1]))
# Usage: python3 run_model.py model_name.pkl scaler_name.pkl SMILE
import pickle, sys
import numpy as np
from rdkit import Chem
from rdkit.Chem import rdMolDescriptors
model = pickle.load(open(sys.argv[1], 'rb'))
scaler = pickle.load(open(sys.argv[2], 'rb'))
compound = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
Chem.MolFromSmiles(sys.argv[3]))
print(model.predict(scaler.transform(np.asarray(compound).reshape(1, -1))))
# Simple benzodiazepine classfication model based on similarity
# to other benzodiazepines
from rdkit import Chem
from rdkit.Chem import rdMolDescriptors
from rdkit import DataStructs
from sklearn.model_selection import train_test_split
data = open('data/benzodiazepine_activator/total_smiles.txt', 'r')
molecules = []
for line in data.readlines():
compound = Chem.MolFromSmiles(line[:-1])
molecules.append(
(rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(compound),
compound))
def model(mol, active_mols):
similarities = []
for active_mol in active_mols:
similarities.append(DataStructs.DiceSimilarity(mol[0], active_mol[0]))
return max(similarities)
train_mols, test_mols = train_test_split(molecules,
test_size=0.1,
random_state=1)
for test_mol in test_mols:
fpdict['ecfp4'] = AllChem.GetMorganFingerprintAsBitVect(smiles, 2, nBits=nbits)
fpdict['ecfp6'] = AllChem.GetMorganFingerprintAsBitVect(smiles, 3, nBits=nbits)
fpdict['fcfp2'] = AllChem.GetMorganFingerprintAsBitVect(smiles,
1,
useFeatures=True,
nBits=nbits)
fpdict['fcfp4'] = AllChem.GetMorganFingerprintAsBitVect(smiles,
2,
useFeatures=True,
nBits=nbits)
fpdict['fcfp6'] = AllChem.GetMorganFingerprintAsBitVect(smiles,
3,
useFeatures=True,
nBits=nbits)
fpdict['maccs'] = MACCSkeys.GenMACCSKeys(smiles)
fpdict['ap'] = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
smiles, nBits=nbits)
fpdict[
'tt'] = rdMolDescriptors.GetHashedTopologicalTorsionFingerprintAsBitVect(
smiles, nBits=nbits)
fpdict['rdk5'] = Chem.RDKFingerprint(smiles,
maxPath=5,
fpSize=nbits,
nBitsPerHash=2)
fpdict['rdk6'] = Chem.RDKFingerprint(smiles,
maxPath=6,
fpSize=nbits,
nBitsPerHash=2)
fpdict['rdk7'] = Chem.RDKFingerprint(smiles,
maxPath=7,
fpSize=nbits,
nBitsPerHash=2)
# Creates a png of a list of SMILES
# Sorts the molecules by taking first molecule and then arranging the rest
# from most to least similar to that molecule
from rdkit.Chem import Draw
from rdkit import Chem
from rdkit import DataStructs
from rdkit.Chem import rdMolDescriptors
data = open('data/benzodiazepine_activator/total_smiles.txt', 'r')
molecules = []
for line in data.readlines():
mol = Chem.MolFromSmiles(line[:-1])
combined = (rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(mol),
mol)
molecules.append(combined)
sorted_mols = sorted(
molecules[1:],
key=lambda x: DataStructs.DiceSimilarity(molecules[0][0], x[0]))
img = Draw.MolsToGridImage([x[1] for x in sorted_mols],
molsPerRow=10,
subImgSize=(200, 200))
img.save('molecules.png')
import rdkit
from rdkit.Chem import AllChem
from rdkit.Chem import MACCSkeys
from rdkit.Chem import rdMolDescriptors
import sys
smi = Chem.SmilesMolSupplier(sys.argv[1], delimiter=',', titleLine=True)
fps = [
AllChem.GetMorganFingerprintAsBitVect(x, 2, useBondTypes=False, nBits=1024)
for x in smi
] ### ECFP4
fps2 = [
AllChem.GetMorganFingerprintAsBitVect(x, 1, useBondTypes=False, nBits=1024)
for x in smi
] ### ECFP2
maccs = [MACCSkeys.GenMACCSKeys(x) for x in smi]
dl = [rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(x) for x in smi]
print("D1,D2,ECFP4,ECFP2,MACCS,DL,AVG,Weighted")
seen = []
for i in range(len(fps)):
d1 = smi[i].GetProp('_Name')
for ii in range(len(fps)):
d2 = smi[ii].GetProp('_Name')
dist = DataStructs.FingerprintSimilarity(fps[i], fps[ii])
dist2 = DataStructs.FingerprintSimilarity(fps2[i], fps2[ii])
distMACCS = DataStructs.FingerprintSimilarity(maccs[i], maccs[ii])
distDL = DataStructs.FingerprintSimilarity(dl[i], dl[ii])
weightedavg = dist * .3 + dist2 * .3 + distDL * .3 + distMACCS * .1
avg = (dist + dist2 + distDL + distMACCS) / 4
print(
str(d1) + "," + str(d2) + "," + str(dist) + "," + str(dist2) +
"," + str(distMACCS) + "," + str(distDL) + "," + str(avg) + "," +
# dictionary
fpFunc_dict = {}
fpFunc_dict['ecfp0'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 0, nBits=nbits)
fpFunc_dict['ecfp2'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 1, nBits=nbits)
fpFunc_dict['ecfp4'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 2, nBits=nbits)
fpFunc_dict['ecfp6'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 3, nBits=nbits)
fpFunc_dict['fcfp2'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 1, useFeatures=True, nBits=nbits)
fpFunc_dict['fcfp4'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 2, useFeatures=True, nBits=nbits)
fpFunc_dict['fcfp6'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 3, useFeatures=True, nBits=nbits)
fpFunc_dict['lecfp4'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 2, nBits=longbits)
fpFunc_dict['lecfp6'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 3, nBits=longbits)
fpFunc_dict['lfcfp4'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 2, useFeatures=True, nBits=longbits)
fpFunc_dict['lfcfp6'] = lambda m: AllChem.GetMorganFingerprintAsBitVect(m, 3, useFeatures=True, nBits=longbits)
fpFunc_dict['maccs'] = lambda m: MACCSkeys.GenMACCSKeys(m)
fpFunc_dict['hashap'] = lambda m: rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(m, nBits=nbits)
fpFunc_dict['hashtt'] = lambda m: rdMolDescriptors.GetHashedTopologicalTorsionFingerprintAsBitVect(m, nBits=nbits)
fpFunc_dict['avalon'] = lambda m: fpAvalon.GetAvalonFP(m, nbits)
fpFunc_dict['laval'] = lambda m: fpAvalon.GetAvalonFP(m, longbits)
fpFunc_dict['rdk5'] = lambda m: Chem.RDKFingerprint(m, maxPath=5, fpSize=nbits, nBitsPerHash=2)
fpFunc_dict['rdk6'] = lambda m: Chem.RDKFingerprint(m, maxPath=6, fpSize=nbits, nBitsPerHash=2)
fpFunc_dict['rdk7'] = lambda m: Chem.RDKFingerprint(m, maxPath=7, fpSize=nbits, nBitsPerHash=2)
fpFunc_dict['tpatf'] = lambda m: get_tpatf(m)
fpFunc_dict['rdkDes'] = lambda m: calc.CalcDescriptors(m)
long_fps = {'laval', 'lecfp4', 'lecfp6', 'lfcfp4', 'lfcfp6'}
fps_to_generate = ['fcfp4', 'rdkDes', 'tpatf', 'rdk5', 'hashap', 'avalon', 'laval', 'rdk7']
ModFileName_LoadedModel_dict = {}
Parameters:
probeMol -- the probe molecule
fpFunction -- the fingerprint function
predictionFunction -- the prediction function of the ML model
kwargs -- additional arguments for drawing
"""
weights = GetAtomicWeightsForModel(probeMol, fpFunction, predictionFunction)
weights, maxWeight = GetStandardizedWeights(weights)
fig = GetSimilarityMapFromWeights(probeMol, weights, **kwargs)
return fig, maxWeight
apDict = {}
apDict['normal'] = lambda m, bits, minl, maxl, bpe, ia: rdMD.GetAtomPairFingerprint(m, minLength=minl, maxLength=maxl, ignoreAtoms=ia)
apDict['hashed'] = lambda m, bits, minl, maxl, bpe, ia: rdMD.GetHashedAtomPairFingerprint(m, nBits=bits, minLength=minl, maxLength=maxl, ignoreAtoms=ia)
apDict['bv'] = lambda m, bits, minl, maxl, bpe, ia: rdMD.GetHashedAtomPairFingerprintAsBitVect(m, nBits=bits, minLength=minl, maxLength=maxl, nBitsPerEntry=bpe, ignoreAtoms=ia)
# usage: lambda m,i: GetAPFingerprint(m, i, fpType, nBits, minLength, maxLength, nBitsPerEntry)
def GetAPFingerprint(mol, atomId=-1, fpType='normal', nBits=2048, minLength=1, maxLength=30, nBitsPerEntry=4):
"""
Calculates the atom pairs fingerprint with the torsions of atomId removed.
Parameters:
mol -- the molecule of interest
atomId -- the atom to remove the pairs for (if -1, no pair is removed)
fpType -- the type of AP fingerprint ('normal', 'hashed', 'bv')
nBits -- the size of the bit vector (only for fpType='bv')
minLength -- the minimum path length for an atom pair
maxLength -- the maxmimum path length for an atom pair
nBitsPerEntry -- the number of bits available for each pair
"""
fpdict["lecfp6"] = lambda m: AllChem.GetMorganFingerprintAsBitVect(
m, 3, nBits=longbits
)
fpdict["lfcfp4"] = lambda m: AllChem.GetMorganFingerprintAsBitVect(
m, 2, useFeatures=True, nBits=longbits
)
fpdict["lfcfp6"] = lambda m: AllChem.GetMorganFingerprintAsBitVect(
m, 3, useFeatures=True, nBits=longbits
)
fpdict["maccs"] = lambda m: MACCSkeys.GenMACCSKeys(m)
fpdict["ap"] = lambda m: Pairs.GetAtomPairFingerprint(m)
fpdict["tt"] = lambda m: Torsions.GetTopologicalTorsionFingerprintAsIntVect(m)
fpdict[
"hashap"
] = lambda m: rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
m, nBits=nbits
)
fpdict[
"hashap_cas_length"
] = lambda m: rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
m, nBits=n_cas_bits
)
fpdict[
"hashtt"
] = lambda m: rdMolDescriptors.GetHashedTopologicalTorsionFingerprintAsBitVect(
m, nBits=nbits
)
fpdict[
"hashtt_cas_length"
] = lambda m: rdMolDescriptors.GetHashedTopologicalTorsionFingerprintAsBitVect(
m, nBits=n_cas_bits
"""
Predict the pKa of an acid from SMILES string
Returns both the fingerprint model prediction and similarity model predictions
"""
from chemical_models import AcidSimilarity, AcidpKa
from rdkit import Chem
from rdkit.Chem import rdMolDescriptors, MACCSkeys
from rdkit.Avalon.pyAvalonTools import GetAvalonFP
import sys
# Load models
sim_model = AcidSimilarity('acid_sim')
fp_model = AcidpKa('pKa_acid')
# Set of acids required for similarity model
acid_data = open('data/pKa/formatted_acidic.txt', 'r')
acids = []
mol = Chem.MolFromSmiles(sys.argv[1])
# Read acids from file
for line in acid_data.readlines():
split = line.split(' ')
acids.append([split[0], float(split[1][:-1]),
rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(Chem.MolFromSmiles(split[0]))])
# Run the models and print results
print("Similarity based model: " + str(sim_model.run(sys.argv[1], acids)))
print("Fingerprint based model: " + str(fp_model.run(GetAvalonFP(mol) + MACCSkeys.GenMACCSKeys(mol) +
rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(mol))))
def pair_fingerprinter(mol):
fp = rdMolDescriptors.GetHashedAtomPairFingerprintAsBitVect(
mol, nBits=fpSize, minLength=minLength, maxLength=maxLength)
return _fp_to_bytes(fp)
import numpy, cPickle
from rdkit import Chem, DataStructs
from rdkit.Chem import rdMolDescriptors as rdmd
# global variables
num_act = 1528
num_dcy = 293606
num_rep = 50
num_percent = 0.1
# fingerprint dictionary
fp_dict = {}
fp_dict['morgan2'] = lambda m: rdmd.GetMorganFingerprintAsBitVect(
m, 2, nBits=1024)
fp_dict['ap'] = lambda m: rdmd.GetHashedAtomPairFingerprintAsBitVect(
m, nBits=2048)
fp_dict['rdk5'] = lambda m: Chem.RDKFingerprint(
m, maxPath=5, fpSize=2048, nBitsPerHash=2)
def getNumpyFP(smiles, fpname, fptype):
m = Chem.MolFromSmiles(smiles)
if m is not None:
# calculate fingerprint
fp = fp_dict[fpname](m)
# convert to numpy array
if fptype == 'bool':
arr = numpy.zeros((1, ), numpy.bool)
elif fptype == 'float':
arr = numpy.zeros((1, ), numpy.float32)
else: