def _transform_mol(self, mol):
"""Private method to transform a skchem molecule.
Use `transform` for the public method, which genericizes the argument
to iterables of mols.
Args:
mol (skchem.Mol): Molecule to calculate fingerprint for.
Returns:
np.array or dict:
Fingerprint as an array (or a dict if sparse).
"""
if self.as_bits and self.n_feats > 0:
fp = GetMorganFingerprintAsBitVect(
mol,
self.radius,
nBits=self.n_feats,
useFeatures=self.use_features,
useBondTypes=self.use_bond_types,
useChirality=self.use_chirality)
res = np.array(0)
ConvertToNumpyArray(fp, res)
res = res.astype(np.uint8)
else:
if self.n_feats <= 0:
res = GetMorganFingerprint(mol,
self.radius,
useFeatures=self.use_features,
useBondTypes=self.use_bond_types,
useChirality=self.use_chirality)
res = res.GetNonzeroElements()
if self.as_bits:
res = {k: int(v > 0) for k, v in res.items()}
else:
res = GetHashedMorganFingerprint(
mol,
self.radius,
nBits=self.n_feats,
useFeatures=self.use_features,
useBondTypes=self.use_bond_types,
useChirality=self.use_chirality)
res = np.array(list(res))
return res
def _one_cats(mol):
""" Function to calculate the CATS pharmacophore descriptor for one molecule.
Descriptions of the individual features can be obtained from the function ``get_cats_sigfactory``.
:param mol: {RDKit molecule} molecule to calculate the descriptor for
:return: {numpy.ndarray} calculated descriptor vector
"""
factory = get_cats_factory()
arr = np.zeros((1,))
ConvertToNumpyArray(Generate.Gen2DFingerprint(mol, factory), arr)
scale = np.array([10 * [sum(arr[i:i + 10])] for i in range(0, 210, 10)]).flatten()
return np.divide(arr, scale, out=np.zeros_like(arr), where=scale != 0).astype('float32')
def _create_fp(smile: AnyStr,
radius: int = 2,
nBits: int = 2048) -> np.ndarray:
atorvation: rdkit.Chem.Mol = Chem.MolFromSmiles(smile)
fingerprint = GetMorganFingerprintAsBitVect(atorvation,
radius=radius,
nBits=nBits)
fp_array: np.ndarray = np.zeros((1, ))
ConvertToNumpyArray(fingerprint, fp_array)
return fp_array
def _rdk2numpy(fps):
""" private function to transform RDKit fingerprints into numpy arrays
:param fps: {list} list of RDKit fingerprints
:return: {numpy.ndarray} fingerprints in array
"""
np_fps = []
for fp in fps:
arr = np.zeros((1,))
ConvertToNumpyArray(fp, arr)
np_fps.append(arr)
return np.array(np_fps).reshape((len(fps), len(np_fps[0])))
def create_mol(df_l, n_bits): # Construct a molecule from a SMILES string # Generate mol column: Returns a Mol object, None on failure. df_l['mol'] = df_l.smiles.apply(Chem.MolFromSmiles) # Create a column for storing the molecular fingerprint as fingerprint object df_l['bv'] = df_l.mol.apply( # Apply the lambda function "calculate_fp" for each molecule lambda x: calculate_fp(x, 'maccs', n_bits) ) # Allocate np.array to hold fp bit-vector (np = numpy) df_l['np_bv'] = np.zeros((len(df_l), df_l['bv'][0].GetNumBits())).tolist() df_l.np_bv = df_l.np_bv.apply(np.array) # Convert the object fingerprint to NumpyArray and store in np_bv df_l.apply(lambda x: ConvertToNumpyArray(x.bv, x.np_bv), axis=1)
def _cats_corr(mols, q):
""" private cats descriptor function to be used in multiprocessing
:param mols: {list/array} molecules (RDKit mol) to calculate the descriptor for
:param q: {queue} multiprocessing queue instance
:return: {numpy.ndarray} calculated descriptor vectors
"""
factory = get_cats_factory()
fps = []
for mol in mols:
arr = np.zeros((1,))
ConvertToNumpyArray(Generate.Gen2DFingerprint(mol, factory), arr)
scale = np.array([10 * [sum(arr[i:i + 10])] for i in range(0, 210, 10)]).flatten()
fps.append(np.divide(arr, scale, out=np.zeros_like(arr), where=scale != 0))
q.put(np.array(fps).reshape((len(mols), 210)).astype('float32'))
def _transform_mol(self, mol):
""" Private method to transform a skchem molecule.
Args:
mol (skchem.Mol): Molecule to calculate fingerprint for.
Returns:
np.array or dict:
Fingerprint as an array (or a dict if sparse).
"""
if self.as_bits and self.n_feats > 0:
fp = GetHashedTopologicalTorsionFingerprintAsBitVect(
mol,
nBits=self.n_feats,
targetSize=self.target_size,
includeChirality=self.use_chirality)
res = np.array(0)
ConvertToNumpyArray(fp, res)
res = res.astype(np.uint8)
else:
if self.n_feats <= 0:
res = GetTopologicalTorsionFingerprint(
mol,
nBits=self.n_feats,
targetSize=self.target_size,
includeChirality=self.use_chirality)
res = res.GetNonzeroElements()
if self.as_bits:
res = {k: int(v > 0) for k, v in res.items()}
else:
res = GetHashedTopologicalTorsionFingerprint(
mol,
nBits=self.n_feats,
targetSize=self.target_size,
includeChirality=self.use_chirality)
res = np.array(list(res))
return res
def to_np(vect, nbits):
arr = numpy.zeros((nbits, ), 'i')
return ConvertToNumpyArray(vect, arr)
def fingerprints_from_mol(cls, mol):
fp = AllChem.GetMorganFingerprint(mol, 3, useFeatures=True)
nfp = np.zeros((1, ))
ConvertToNumpyArray(fp, nfp)
return nfp
def numpy(self):
np_ = np.zeros((1, ))
ConvertToNumpyArray(self.fp, np_)
return np_
def get_bit_vector(fp):
arr = np.zeros((1,))
ConvertToNumpyArray(fp, arr)
return(arr)