def pred_multiple_prod(testdf, model, path_vec, k=1):
"""Predicts the products of specific chemical reactions.
Input is reactant SMILES strings.
The default predicted consequence is one, you can change the value of k to get more probable forecasted results.
Args:
"""
a = []
b = []
c = []
for i in range(len(testdf)):
smi = testdf['Reactants'][i]
std = transform_r2p_str(smi, model, path_vec, k)
c.append(std)
[a.append(std[i]) for i in range(len(std))]
for j in range(len(std)):
col = 'Product'
b.append(col)
out = pd.DataFrame(data=c, columns=b)
out.insert(
0,
'Reactants',
testdf['Reactants'].values,
)
df = struc2mol(pd.DataFrame(data=a, columns=['smiles']))
display(
PandasTools.FrameToGridImage(df,
column='mol',
legendsCol='smiles',
molsPerRow=5))
return out
def output_single_prod(smi, model, path_vec, k=15):
"""A function used to predict the product of a specific chemical reactions with the input of reactant smiles string.
When using beamsearch, the value of k is 15.
"""
a = ['Reactant', 'Product']
b = []
c = [smi]
std = transform_r2p_str(smi, model, path_vec, k)
for j in range(len(std)):
if std[j] == smi.upper(
): # still need some more work, not applied to all reactions
prd = std[j]
break
elif smi.replace('#', '') == std[j]:
prd = std[j]
break
else:
prd = std[14]
b.append(prd)
c.append(prd)
out = pd.DataFrame(data=b, columns=['Product'])
out.insert(0, 'Reactant', smi)
df = struc2mol(pd.DataFrame(data=c, columns=['smiles']))
df.insert(3, 'legend', a)
display(
PandasTools.FrameToGridImage(df,
column='mol',
legendsCol='legend',
molsPerRow=5))
return out
def grid_image(
df,
filename,
molobj=True,
smi='smiles'): # list of molecules to print and substructre to align
"""
Creates and saves grid image of 2D drawings of molecules.
Accepts dataframe containing a column titled "Molecule" that contains RDKit molecule objects.
Accepts filename as string (without .png) for image file.
Returns nothing, saves file in current directory.
_____________________________
Keyword Arguments:
molobj=True, if RDKit MolObj column exists in df. (Must be headed "Molecule")
smi='smiles', if molojb=False then use column titled smi to create MolObj column.
"""
if not molobj: # no molobj exists
PandasTools.AddMoleculeColumnToFrame(df,
smi,
'Molecule',
includeFingerprints=True)
# this code makes multiple images of n molecules. May be prefered for large sets of molecules.
# create images of molecules in dataframe
mol_image = PandasTools.FrameToGridImage(
df,
column='Molecule',
molsPerRow=3,
subImgSize=(800, 400),
legends=[str(i + 1) for i in range(len(df['Molecule']))])
mol_image.save(
filename +
'.png') # shold use a better naming scheme to avoid overwrites.
def run(self, k=15):
#self.clear
smi = self.sms
model = load_model()
# path_vec = load_path_vector()
path_vec = self.path_vec
a = ['Reactant', 'Product']
b = []
c = [smi]
std = tranform(smi, model, path_vec, k)
for j in range(len(std)):
if std[j] == smi.upper():
prd = std[j]
break
elif smi.replace('#', '') == std[j]:
prd = std[j]
break
else:
prd = std[14]
b.append(prd)
c.append(prd)
out = pd.DataFrame(data=b, columns=['Product'])
out.insert(0, 'Reactant', smi)
df = struc2mol(pd.DataFrame(data=c, columns=['smiles']))
df.insert(3, 'legend', a)
out = (PandasTools.FrameToGridImage(df,
column='mol',
legendsCol='smiles',
molsPerRow=2))
a = np.array(out)
scipy.misc.imsave('outfile.png', a)
figname = 'outfile.png'
#time.sleep(3)
self._signal.emit(str(figname))
def test_FrameToGridImage(self):
# This test only makes sure that we get no exception. To see the created images, set
# interactive to True
interactive = False
self.assertTrue(True)
df = self.df
result = PandasTools.FrameToGridImage(df)
if interactive:
result.show()
result = PandasTools.FrameToGridImage(
df, legendsCol='PUBCHEM_IUPAC_INCHIKEY')
if interactive:
result.show()
result = PandasTools.FrameToGridImage(df, legendsCol=df.index.name)
if interactive:
result.show()
def test_force_Kekulize():
df = pd.read_csv('All_Moles_Tested_Data.csv')
i= 0
mol_list = []
for smile in df['smiles']:
mol = MolFromSmiles(smile)
x = find_custom_Kekulize_set(smile, max_atoms= 60, max_degree= 5,printMe = False)
for index in x:
mol.GetAtomWithIdx(index).SetAtomicNum(32)
mol_list.append(mol)
df['mol'] = pd.DataFrame({'mol':mol_list})
unit = 5
for i in range(0,len(df)//unit):
display(PandasTools.FrameToGridImage(df.iloc[i*unit:i*unit+unit],column='mol', legendsCol='',molsPerRow=unit))
if((len(df)%unit>0)*1):
display(PandasTools.FrameToGridImage(df.iloc[len(df)//unit*unit:len(df)],column='mol', legendsCol='',molsPerRow=unit))
def output_grid(df, imagename):
#use PandasTools.FrameToGridImage
template = Chem.MolFromSmarts(smartsq)
AllChem.Compute2DCoords(template)
df['mol_strip'].apply(
lambda x: AllChem.GenerateDepictionMatching2DStructure(x, template))
df['matches'] = df['mol_strip'].apply(
lambda x: x.GetSubstructMatch(template))
frameimage = PandasTools.FrameToGridImage(df,
column='mol_strip',
legendsCol='molregno',
molsPerRow=10,
highlightAtomLists=list(
df['matches']))
frameimage.save(imagename)
def pred_single_prod(smi, model, path_vec, k=1):
"""A function used to predict the product of a specific chemical reactions with the input of reactant smiles string.
The default predicted consequence is one, you can change the value of k to get more probable forecasted results.
"""
c = []
b = []
std = transform_r2p_str(smi, model, path_vec, k)
c.append(std)
for j in range(len(std)):
col = 'Product'
b.append(col)
out = pd.DataFrame(data=c, columns=b)
out.insert(0, 'Reactant', smi)
df = struc2mol(pd.DataFrame(data=std, columns=['smiles']))
display(
PandasTools.FrameToGridImage(df,
column='mol',
legendsCol='smiles',
molsPerRow=5))
return out
def output_multiple_prod(testdf, model, path_vec, k=15):
"""A function used to output the product of many specific chemical reactions with the input of reactant smiles strings.
The default value for k is 15.
"""
a = []
b = []
c = []
for i in range(len(testdf)):
smi = testdf['Reactants'][i]
a.append('Reactant')
c.append(smi)
std = transform_r2p_str(smi, model, path_vec, k)
for j in range(len(std)):
if std[j] == smi.upper():
prd = std[j]
break
elif smi.replace('#', '') == std[j]:
prd = std[j]
break
else:
prd = std[14]
a.append('Product')
c.append(prd)
b.append(prd)
out = pd.DataFrame(data=b, columns=['Products'])
out.insert(
0,
'Reactants',
testdf['Reactants'].values,
)
df = struc2mol(pd.DataFrame(data=c, columns=['smiles']))
df.insert(3, 'legend', a)
display(
PandasTools.FrameToGridImage(df,
column='mol',
legendsCol='legend',
molsPerRow=2))
return out
def test_github2380(self):
from rdkit.Chem.Draw import IPythonConsole
IPythonConsole.ipython_useSVG = True
df = PandasTools.LoadSDF(getStreamIO(methane + peroxide))
_ = PandasTools.FrameToGridImage(df)
'SUBSTANCE_SYNONYM', 'SUBSTANCE_COMMENT', 'SUBSTANCE_URL',
'INCHIKEY_1.05_CHEMAXON_19.27.0'
]]
# Display table
# thesis_df
# fixes mol display in dataframes (RDKit Issue# 2673)
from IPython.display import HTML
HTML(thesis_df.head(len(thesis_df.index)).to_html())
# In[7]:
# we can also display just the molecules like this:
PandasTools.FrameToGridImage(thesis_df,
column='RDMol',
molsPerRow=2,
subImgSize=(400, 400),
legendsCol="DATASOURCE_REGID")
# In[8]:
# Now we need to caluclate the InChIs from RDKit and add to thesis_df
# These are InChI 1.05 as computed by RDKit 2019.09.2 release.
inchi_list = []
for mol in thesis_df['RDMol']:
inchi = Chem.MolToInchi(mol)
inchi_list.append(inchi)
# add to dataframe
thesis_df['INCHI_1.05_RDKIT_2019.09.2'] = inchi_list
