Desc2DTransformer: RDKit descriptors transformer¶

The descriptors transformer can convert molecules into a list of RDKit descriptors. It largely follows the API of the other transformers, but has a few extra methods and properties to manage the descriptors.

In [1]:

from rdkit import Chem
import numpy as np
import matplotlib.pyplot as plt
from scikit_mol.descriptors import MolecularDescriptorTransformer

from rdkit import Chem import numpy as np import matplotlib.pyplot as plt from scikit_mol.descriptors import MolecularDescriptorTransformer

After instantiation of the descriptor transformer, we can query which descriptors it found available in the RDKit framework.

In [2]:

descriptor = MolecularDescriptorTransformer()
available_descriptors = descriptor.available_descriptors
print(f"There are {len(available_descriptors)} available descriptors")
print(f"The first five descriptor names: {available_descriptors[:5]}")

descriptor = MolecularDescriptorTransformer() available_descriptors = descriptor.available_descriptors print(f"There are {len(available_descriptors)} available descriptors") print(f"The first five descriptor names: {available_descriptors[:5]}")

There are 217 available descriptors
The first five descriptor names: ['MaxAbsEStateIndex', 'MaxEStateIndex', 'MinAbsEStateIndex', 'MinEStateIndex', 'qed']

We can transform molecules to their descriptor profiles

In [3]:

smiles_list = ["CCCC", "c1ccccc1"]
mols = [Chem.MolFromSmiles(smiles) for smiles in smiles_list]

features = descriptor.transform(mols)
_ = plt.plot(np.array(features).T)

smiles_list = ["CCCC", "c1ccccc1"] mols = [Chem.MolFromSmiles(smiles) for smiles in smiles_list] features = descriptor.transform(mols) _ = plt.plot(np.array(features).T)

If we only want some of them, this can be specified at object instantiation.

In [4]:

some_descriptors = MolecularDescriptorTransformer(
    desc_list=["HeavyAtomCount", "FractionCSP3", "RingCount", "MolLogP", "MolWt"]
)
print(f"Selected descriptors are {some_descriptors.selected_descriptors}")
features = some_descriptors.transform(mols)

some_descriptors = MolecularDescriptorTransformer( desc_list=["HeavyAtomCount", "FractionCSP3", "RingCount", "MolLogP", "MolWt"] ) print(f"Selected descriptors are {some_descriptors.selected_descriptors}") features = some_descriptors.transform(mols)

Selected descriptors are ['HeavyAtomCount', 'FractionCSP3', 'RingCount', 'MolLogP', 'MolWt']

If we want to update the selected descriptors on an already existing object, this can be done via the .set_params() method

In [5]:

print(
    some_descriptors.set_params(
        desc_list=["HeavyAtomCount", "FractionCSP3", "RingCount"]
    )
)

print( some_descriptors.set_params( desc_list=["HeavyAtomCount", "FractionCSP3", "RingCount"] ) )

MolecularDescriptorTransformer(desc_list=['HeavyAtomCount', 'FractionCSP3',
                                          'RingCount'])

In [ ]: