def test_single_carbon(self):
"""Test that single carbon atom is featurized properly."""
raw_smiles = ['C']
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mol_list = featurizer.featurize(mols)
mol = mol_list[0]
# Only one carbon
assert mol.get_num_atoms() == 1
# No bonds, so degree adjacency lists are empty
deg_adj_lists = mol.get_deg_adjacency_lists()
assert np.array_equal(deg_adj_lists[0],
np.zeros([1,0], dtype=np.int32))
assert np.array_equal(deg_adj_lists[1],
np.zeros([0,1], dtype=np.int32))
assert np.array_equal(deg_adj_lists[2],
np.zeros([0,2], dtype=np.int32))
assert np.array_equal(deg_adj_lists[3],
np.zeros([0,3], dtype=np.int32))
assert np.array_equal(deg_adj_lists[4],
np.zeros([0,4], dtype=np.int32))
assert np.array_equal(deg_adj_lists[5],
np.zeros([0,5], dtype=np.int32))
assert np.array_equal(deg_adj_lists[6],
np.zeros([0,6], dtype=np.int32))
def test_alkane(self):
"""Test on simple alkane"""
raw_smiles = ['CCC']
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mol_list = featurizer.featurize(mols)
mol = mol_list[0]
# 3 carbonds in alkane
assert mol.get_num_atoms() == 3
deg_adj_lists = mol.get_deg_adjacency_lists()
assert np.array_equal(deg_adj_lists[0],
np.zeros([0,0], dtype=np.int32))
# Outer two carbonds are connected to central carbon
assert np.array_equal(deg_adj_lists[1],
np.array([[2], [2]], dtype=np.int32))
# Central carbon connected to outer two
assert np.array_equal(deg_adj_lists[2],
np.array([[0,1]], dtype=np.int32))
assert np.array_equal(deg_adj_lists[3],
np.zeros([0,3], dtype=np.int32))
assert np.array_equal(deg_adj_lists[4],
np.zeros([0,4], dtype=np.int32))
assert np.array_equal(deg_adj_lists[5],
np.zeros([0,5], dtype=np.int32))
assert np.array_equal(deg_adj_lists[6],
np.zeros([0,6], dtype=np.int32))
def test_graph_gather(self):
"""Test that GraphGather can be invoked."""
batch_size = 2
n_features = 75
n_atoms = 4 # In CCC and C, there are 4 atoms
raw_smiles = ['CCC', 'C']
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mols = featurizer.featurize(mols)
multi_mol = ConvMol.agglomerate_mols(mols)
atom_features = multi_mol.get_atom_features()
degree_slice = multi_mol.deg_slice
membership = multi_mol.membership
deg_adjs = multi_mol.get_deg_adjacency_lists()[1:]
with self.test_session() as sess:
atom_features = tf.convert_to_tensor(atom_features,
dtype=tf.float32)
degree_slice = tf.convert_to_tensor(degree_slice, dtype=tf.int32)
membership = tf.convert_to_tensor(membership, dtype=tf.int32)
deg_adjs_tf = []
for deg_adj in deg_adjs:
deg_adjs_tf.append(
tf.convert_to_tensor(deg_adj, dtype=tf.int32))
args = [atom_features, degree_slice, membership] + deg_adjs_tf
out_tensor = GraphGather(batch_size)(*args)
sess.run(tf.global_variables_initializer())
out_tensor = out_tensor.eval()
# TODO(rbharath): Why is it 2*n_features instead of n_features?
assert out_tensor.shape == (batch_size, 2 * n_features)
def test_carbon_nitrogen(self):
"""Test on carbon nitrogen molecule"""
# Note there is a central carbon of degree 4, with 3 carbons and
# one nitrogen of degree 1 (connected only to central carbon).
raw_smiles = ['C[N+](C)(C)C']
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mols = featurizer.featurize(mols)
mol = mols[0]
# 5 atoms in compound
assert mol.get_num_atoms() == 5
# Get the adjacency lists grouped by degree
deg_adj_lists = mol.get_deg_adjacency_lists()
assert np.array_equal(deg_adj_lists[0],
np.zeros([0,0], dtype=np.int32))
# The 4 outer atoms connected to central carbon
assert np.array_equal(deg_adj_lists[1],
np.array([[4], [4], [4], [4]], dtype=np.int32))
assert np.array_equal(deg_adj_lists[2],
np.zeros([0,2], dtype=np.int32))
assert np.array_equal(deg_adj_lists[3],
np.zeros([0,3], dtype=np.int32))
# Central carbon connected to everything else.
assert np.array_equal(deg_adj_lists[4],
np.array([[0, 1, 2, 3]], dtype=np.int32))
assert np.array_equal(deg_adj_lists[5],
np.zeros([0,5], dtype=np.int32))
assert np.array_equal(deg_adj_lists[6],
np.zeros([0,6], dtype=np.int32))
def test_carbon_nitrogen(self):
"""Test on carbon nitrogen molecule"""
# Note there is a central carbon of degree 4, with 3 carbons and
# one nitrogen of degree 1 (connected only to central carbon).
raw_smiles = ['C[N+](C)(C)C']
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mols = featurizer.featurize(mols)
mol = mols[0]
# 5 atoms in compound
assert mol.get_num_atoms() == 5
# Get the adjacency lists grouped by degree
deg_adj_lists = mol.get_deg_adjacency_lists()
assert np.array_equal(deg_adj_lists[0], np.zeros([0, 0], dtype=np.int32))
# The 4 outer atoms connected to central carbon
assert np.array_equal(deg_adj_lists[1],
np.array([[4], [4], [4], [4]], dtype=np.int32))
assert np.array_equal(deg_adj_lists[2], np.zeros([0, 2], dtype=np.int32))
assert np.array_equal(deg_adj_lists[3], np.zeros([0, 3], dtype=np.int32))
# Central carbon connected to everything else.
assert np.array_equal(deg_adj_lists[4],
np.array([[0, 1, 2, 3]], dtype=np.int32))
assert np.array_equal(deg_adj_lists[5], np.zeros([0, 5], dtype=np.int32))
assert np.array_equal(deg_adj_lists[6], np.zeros([0, 6], dtype=np.int32))
def test_alkane(self):
"""Test on simple alkane"""
raw_smiles = ['CCC']
import rdkit.Chem
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mol_list = featurizer.featurize(mols)
mol = mol_list[0]
# 3 carbonds in alkane
assert mol.get_num_atoms() == 3
deg_adj_lists = mol.get_deg_adjacency_lists()
assert np.array_equal(deg_adj_lists[0], np.zeros([0, 0],
dtype=np.int32))
# Outer two carbonds are connected to central carbon
assert np.array_equal(deg_adj_lists[1],
np.array([[2], [2]], dtype=np.int32))
# Central carbon connected to outer two
assert np.array_equal(deg_adj_lists[2],
np.array([[0, 1]], dtype=np.int32))
assert np.array_equal(deg_adj_lists[3], np.zeros([0, 3],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[4], np.zeros([0, 4],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[5], np.zeros([0, 5],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[6], np.zeros([0, 6],
dtype=np.int32))
def test_single_carbon(self):
"""Test that single carbon atom is featurized properly."""
raw_smiles = ['C']
import rdkit
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mol_list = featurizer.featurize(mols)
mol = mol_list[0]
# Only one carbon
assert mol.get_num_atoms() == 1
# No bonds, so degree adjacency lists are empty
deg_adj_lists = mol.get_deg_adjacency_lists()
assert np.array_equal(deg_adj_lists[0], np.zeros([1, 0],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[1], np.zeros([0, 1],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[2], np.zeros([0, 2],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[3], np.zeros([0, 3],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[4], np.zeros([0, 4],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[5], np.zeros([0, 5],
dtype=np.int32))
assert np.array_equal(deg_adj_lists[6], np.zeros([0, 6],
dtype=np.int32))
def test_graph_gather(self):
"""Test that GraphGather can be invoked."""
batch_size = 2
n_features = 75
n_atoms = 4 # In CCC and C, there are 4 atoms
raw_smiles = ['CCC', 'C']
mols = [rdkit.Chem.MolFromSmiles(s) for s in raw_smiles]
featurizer = ConvMolFeaturizer()
mols = featurizer.featurize(mols)
multi_mol = ConvMol.agglomerate_mols(mols)
atom_features = multi_mol.get_atom_features()
degree_slice = multi_mol.deg_slice
membership = multi_mol.membership
deg_adjs = multi_mol.get_deg_adjacency_lists()[1:]
with self.session() as sess:
atom_features = tf.convert_to_tensor(atom_features, dtype=tf.float32)
degree_slice = tf.convert_to_tensor(degree_slice, dtype=tf.int32)
membership = tf.convert_to_tensor(membership, dtype=tf.int32)
deg_adjs_tf = []
for deg_adj in deg_adjs:
deg_adjs_tf.append(tf.convert_to_tensor(deg_adj, dtype=tf.int32))
args = [atom_features, degree_slice, membership] + deg_adjs_tf
out_tensor = GraphGather(batch_size)(*args)
sess.run(tf.global_variables_initializer())
out_tensor = out_tensor.eval()
# TODO(rbharath): Why is it 2*n_features instead of n_features?
assert out_tensor.shape == (batch_size, 2 * n_features)
def test_per_atom_fragmentation(self):
"""checks if instantiating featurizer with per_atom_fragmentation=True
leads to as many fragments' features, as many atoms mol has for any mol"""
import rdkit.Chem
raw_smiles = ['CC(CO)Cc1ccccc1', 'CC']
mols = [rdkit.Chem.MolFromSmiles(m) for m in raw_smiles]
featurizer = ConvMolFeaturizer(per_atom_fragmentation=True)
feat = featurizer.featurize(mols)
for i, j in zip(feat, mols):
assert len(i) == j.GetNumHeavyAtoms()
def predict_on_smiles(self, smiles, transformers=[], untransform=False):
"""Generates predictions on a numpy array of smile strings
# Returns:
y_: numpy ndarray of shape (n_samples, n_tasks)
"""
max_index = len(smiles) - 1
n_tasks = len(self.outputs)
num_batches = (max_index // self.batch_size) + 1
featurizer = ConvMolFeaturizer()
y_ = []
for i in range(num_batches):
start = i * self.batch_size
end = min((i + 1) * self.batch_size, max_index + 1)
smiles_batch = smiles[start:end]
y_.append(
self.predict_on_smiles_batch(smiles_batch, featurizer, transformers))
y_ = np.concatenate(y_, axis=0)[:max_index + 1]
y_ = y_.reshape(-1, n_tasks)
if untransform:
y_ = undo_transforms(y_, transformers)
return y_
"""
import warnings
warnings.filterwarnings('ignore')
import deepchem as dc
#from deepchem.models.tensorgraph.models.graph_models import MPNNTensorGraph
from deepchem.models.tensorgraph.models.graph_models import GraphConvModel
#from deepchem.feat import WeaveFeaturizer
from deepchem.feat.graph_features import ConvMolFeaturizer
from deepchem.feat.graph_features import WeaveFeaturizer
from deepchem.data.data_loader import CSVLoader
import pandas as pd
import numpy as np
featurizer = ConvMolFeaturizer()
#featurizer = WeaveFeaturizer(graph_distance=True, explicit_H=False)
train_loader = CSVLoader(tasks=['LogD7.4'],
smiles_field='smiles',
featurizer=featurizer)
test_loader = CSVLoader(tasks=['LogD7.4'],
smiles_field='smiles',
featurizer=featurizer)
X_train = train_loader.featurize('../demo_data/reg/training_set.csv')
X_test = test_loader.featurize('../demo_data/reg/testing_set.csv')
model = GraphConvModel(n_tasks=1, mode='regression')
model.fit(X_train)
print(model.predict(X_test))
