def get_descriptors(df):
PandasTools.ChangeMoleculeRendering(renderer='String')
Lmol = df['ROMol']
Ldescriptors = []
for m in Lmol:
# Calculer les propriétés chimiques
MW = round(Descriptors.ExactMolWt(m), 1)
LogP = round(Descriptors.MolLogP(m), 1)
TPSA = round(Descriptors.TPSA(m), 1)
LabuteASA = round(Descriptors.LabuteASA(m), 1)
HBA = Descriptors.NumHAcceptors(m)
HBD = Descriptors.NumHDonors(m)
FCSP3 = Lipinski.FractionCSP3(m)
MQN8 = rdMolDescriptors.MQNs_(m)[7]
MQN10 = rdMolDescriptors.MQNs_(m)[9]
NAR = Lipinski.NumAromaticRings(m)
NRB = Chem.Descriptors.NumRotatableBonds(m)
Ldescriptors.append([
MW, LogP, TPSA, LabuteASA, HBA, HBD, FCSP3, MQN8, MQN10, NAR, NRB
])
# Create pandas row for conditions results with values and information whether rule of five is violated
prop_df = pd.DataFrame(Ldescriptors)
prop_df.columns = [
'MW', 'LogP', 'TPSA', 'LabuteASA', 'HBA', 'HBD', 'FCSP3', 'MQN8',
'MQN10', 'NAR', 'NRB'
]
prop_df = prop_df.set_index(df.index)
return prop_df
def get_frags_number(df):
PandasTools.ChangeMoleculeRendering(renderer='String')
New_column = pd.DataFrame({
'Frags_number':
[len(Chem.rdmolops.GetMolFrags(mol)) for mol in df['ROMol']]
})
New_column = New_column.set_index(df.index)
return New_column
def get_data(self):
self.data['Molecule'] = self.molecules
self.data['Reference'] = self.reference
df = pd.DataFrame(data=self.data)
PandasTools.ChangeMoleculeRendering(df)
# reinitialize instance
self.__init__()
return df
def test_AddMoleculeColumnToFrame(self):
df = PandasTools.LoadSDF(getStreamIO(methane + peroxide),
isomericSmiles=True,
smilesName='Smiles')
PandasTools.ChangeMoleculeRendering(frame=df, renderer='String')
del df['ROMol']
self.assertNotIn('ROMol', str(df))
PandasTools.AddMoleculeColumnToFrame(df, includeFingerprints=False)
self.assertIn('ROMol', str(df))
def show_decomp(mols, cores, options, item=False):
decomp = do_decomp(mols, cores, options)
if item:
rows = decomp.GetRGroupsAsRows()
items = [
'{}:{}'.format(group, Chem.MolToSmiles(row[group])) for row in rows
for group in row
]
return ' '.join(items)
else:
cols = decomp.GetRGroupsAsColumns()
cols['mol'] = mols
cols['input core'] = cores[0]
df = pd.DataFrame(cols)
PandasTools.ChangeMoleculeRendering(df)
return HTML(df.to_html())
def toDataFrame(self, fields=None, molAsImage=True, sketch=True):
"""
Returns a pandas.DataFrame of the SmallMolLib object.
Parameters
----------
fields: list
The list of fields to convert into a pandas DataFrame column
molAsImage: bool
If True, the rdkit.Chem.rdchem.Mol is converted into an image
sketch: bool
If True, the molecule are rendered to be 2D
Returns
-------
dataframe: pandas.DataFrame
The pandas DataFrame
"""
from rdkit.Chem.AllChem import Compute2DCoords
import pandas as pd
from copy import deepcopy
if fields is not None:
if not isinstance(fields, list):
raise TypeError(
f"The argument fields passed {type(fields)} should be a list "
)
else:
fields = ["ligname", "_mol"] if molAsImage else ["ligname"]
records = []
indexes = []
for i, m in enumerate(self._mols):
row = dict((f, m.__getattribute__(f)) for f in fields)
if sketch:
mm = deepcopy(m._mol)
Compute2DCoords(mm)
row["_mol"] = mm
records.append(row)
indexes.append(i)
df = pd.DataFrame(records, columns=fields, index=indexes)
if molAsImage:
PandasTools.ChangeMoleculeRendering(df)
return df
def setUp(self):
PandasTools.InstallPandasTools()
PandasTools.ChangeMoleculeRendering(renderer='PNG')
PandasTools.pd.set_option('display.max_columns', None)
self._molRepresentation = PandasTools.molRepresentation
self._highlightSubstructures = PandasTools.highlightSubstructures
def setUp(self):
PandasTools.InstallPandasTools()
PandasTools.ChangeMoleculeRendering(renderer='PNG')
self._molRepresentation = PandasTools.molRepresentation
self._highlightSubstructures = PandasTools.highlightSubstructures
