def load_data(num):
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
# atom_featurizer = WeaveAtomFeaturizer()
# bond_featurizer = WeaveEdgeFeaturizer()
# atom_featurizer = node_featurizer
# bond_featurizer = edge_featurizer
# m = 'CCO'
# mol = Chem.MolFromSmiles(m)
# num_bonds = mol.GetNumBonds()
# print(mol,num_bonds)
# smiles_to_bigraph(m, add_self_loop=False, node_featurizer=atom_featurizer,edge_featurizer=bond_featurizer)
# print(bond_featurizer(mol)['distance'].shape)
# print(atom_featurizer(mol)['node_type'].shape)
trainmols, train_y = read_from_rdkit(num,0)
testmols, test_y = read_from_rdkit(num,2)
train_g = [smiles_to_bigraph(m, add_self_loop=False, node_featurizer=atom_featurizer,edge_featurizer=bond_featurizer) for m in trainmols]
train_y = np.array(train_y, dtype=np.int64)
train_y = OneHotEncoder(sparse=False).fit(train_y.reshape(-1,1)).transform(train_y.reshape(-1,1))
train_y = np.array(train_y, dtype=np.float32)
print("Training set ",len(train_g))
test_g = [smiles_to_bigraph(m, add_self_loop=False, node_featurizer=atom_featurizer,edge_featurizer=bond_featurizer) for m in testmols]
test_y = np.array(test_y, dtype=np.int64)
test_y = OneHotEncoder(sparse=False).fit(test_y.reshape(-1,1)).transform(test_y.reshape(-1,1))
test_y = np.array(test_y, dtype=np.float32)
print("Test set",len(test_g))
print("Data loaded.")
return train_g, train_y, test_g, test_y
def main():
"""
:param n_trials: int specifying number of random train/test splits to use
:param test_set_size: float in range [0, 1] specifying fraction of dataset to use as test set
"""
mpi_comm = MPI.COMM_WORLD
mpi_rank = mpi_comm.Get_rank()
mpi_size = mpi_comm.Get_size()
df = pd.read_csv('data/sars_lip.csv')
smiles_list = df['smiles']
my_border_low, my_border_high = return_borders(mpi_rank, len(smiles_list), mpi_size)
my_smiles = smiles_list[my_border_low:my_border_high]
my_mols = np.array([Chem.MolFromSmiles(m) for m in my_smiles])
# Initialise featurisers
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
e_feats = bond_featurizer.feat_size('e')
n_feats = atom_featurizer.feat_size('h')
my_graphs = np.array([mol_to_bigraph(m, node_featurizer=atom_featurizer,
edge_featurizer=bond_featurizer) for m in my_mols])
sendcounts = np.array(mpi_comm.gather(len(my_graphs), root=0))
# my_descs = np.array([generate_descriptors(m) for m in my_smiles])
# if mpi_rank == 0:
# descs = np.empty((len(smiles_list), 114), dtype=np.float64)
# else:
# descs = None
# mpi_comm.Gatherv(sendbuf=my_descs, recvbuf=(descs, sendcounts), root=0)
graphs = mpi_comm.gather(my_graphs, root=0)
X = graphs[0]
if mpi_rank==0:
for graph in graphs:
X = X.vstack([X,graph])
# np.save('/rds-d2/user/wjm41/hpc-work/sars_descs.npy', descs)
np.save('/rds-d2/user/wjm41/hpc-work/sars_graphs.npy', X)
print('SAVED!')
def __CanonicalFeatureize(self, train_mols) -> list:
atom_featurizer = CanonicalAtomFeaturizer("n_feat")
bond_featurizer = CanonicalBondFeaturizer("e_feat")
train_graph = [
mol_to_bigraph(mol,
node_featurizer=atom_featurizer,
edge_featurizer=bond_featurizer)
for mol in train_mols
]
return train_graph
def mol_to_graph(mols: list, canonical: bool = False) -> dgl.DGLGraph:
if canonical:
graph = [
mol_to_bigraph(
mol,
node_featurizer=CanonicalAtomFeaturizer(),
edge_featurizer=CanonicalBondFeaturizer(),
)
for mol in mols
]
else:
graph = [mol_to_bigraph(m, node_featurizer=MyNodeFeaturizer()) for m in mols]
return graph
def moonshot():
from dgllife.utils import mol_to_bigraph, CanonicalAtomFeaturizer
import pandas as pd
import os
df = pd.read_csv(
os.path.dirname(graca.data.collections.__file__) +
"/covid_submissions_all_info.csv")
df = df.dropna(subset=["f_avg_pIC50"])
from rdkit import Chem
from rdkit.Chem import MCS
ds = []
for idx0, row0 in df.iterrows():
smiles0 = row0["SMILES"]
mol0 = Chem.MolFromSmiles(smiles0)
for idx1, row1 in df.iloc[idx0 + 1:].iterrows():
smiles1 = row1["SMILES"]
mol1 = Chem.MolFromSmiles(smiles1)
res = MCS.FindMCS([mol0, mol1])
if res.numAtoms > 15:
ds.append((
mol_to_bigraph(mol1,
node_featurizer=CanonicalAtomFeaturizer(
atom_data_field='feat')),
mol_to_bigraph(mol0,
node_featurizer=CanonicalAtomFeaturizer(
atom_data_field='feat')),
row1["f_avg_pIC50"],
row0["f_avg_pIC50"],
))
ds_tr = ds[:500]
ds_te = ds[500:]
return ds_tr, ds_te
def __init__(
self,
smiles: str,
g: Union[dgl.DGLGraph, None] = None,
y: Union[float, None] = None,
extra: dict = {},
featurizer: callable = functools.partial(
smiles_to_bigraph,
node_featurizer=CanonicalAtomFeaturizer(atom_data_field="h"),
),
) -> None:
self.smiles = smiles
self.g = g
self.y = y
self.extra = extra
self.featurizer = featurizer
def init_featurizer(args):
"""Initialize node/edge featurizer
Parameters
----------
args : dict
Settings
Returns
-------
args : dict
Settings with featurizers updated
"""
if args['model'] in [
'gin_supervised_contextpred', 'gin_supervised_infomax',
'gin_supervised_edgepred', 'gin_supervised_masking'
]:
from dgllife.utils import PretrainAtomFeaturizer, PretrainBondFeaturizer
args['atom_featurizer_type'] = 'pre_train'
args['bond_featurizer_type'] = 'pre_train'
args['node_featurizer'] = PretrainAtomFeaturizer()
args['edge_featurizer'] = PretrainBondFeaturizer()
return args
if args['atom_featurizer_type'] == 'canonical':
from dgllife.utils import CanonicalAtomFeaturizer
args['node_featurizer'] = CanonicalAtomFeaturizer()
elif args['atom_featurizer_type'] == 'attentivefp':
from dgllife.utils import AttentiveFPAtomFeaturizer
args['node_featurizer'] = AttentiveFPAtomFeaturizer()
else:
return ValueError(
"Expect node_featurizer to be in ['canonical', 'attentivefp'], "
"got {}".format(args['atom_featurizer_type']))
if args['model'] in ['Weave', 'MPNN', 'AttentiveFP']:
if args['bond_featurizer_type'] == 'canonical':
from dgllife.utils import CanonicalBondFeaturizer
args['edge_featurizer'] = CanonicalBondFeaturizer()
elif args['bond_featurizer_type'] == 'attentivefp':
from dgllife.utils import AttentiveFPBondFeaturizer
args['edge_featurizer'] = AttentiveFPBondFeaturizer()
else:
args['edge_featurizer'] = None
return args
def _make_atom_graph(self,
pdb_code=None,
pdb_path=None,
node_featurizer=None,
edge_featurizer=None,
graph_type='bigraph'):
"""
Create atom-level graph from PDB structure
:param graph_type:
:param pdb_code:
:param pdb_path:
:param node_featurizer:
:param edge_featurizer:
:return:
"""
if node_featurizer is None:
node_featurizer = CanonicalAtomFeaturizer()
if edge_featurizer is None:
edge_featurizer = CanonicalBondFeaturizer()
# Read in protein as mol
# if pdb_path:
if pdb_code:
pdb_path = self.pdb_dir + pdb_code + '.pdb'
if not os.path.isfile(pdb_path):
self._download_pdb(pdb_code)
assert os.path.isfile(pdb_path)
mol = MolFromPDBFile(pdb_path)
# DGL mol to graph
if graph_type == 'bigraph':
g = mol_to_bigraph(mol,
node_featurizer=node_featurizer,
edge_featurizer=edge_featurizer)
elif graph_type == 'complete':
g = mol_to_complete_graph(
mol,
node_featurizer=node_featurizer,
)
elif graph_type == 'k_nn':
raise NotImplementedError
print(g)
return g
def __init__(self,
smiles,
y,
node_enc=CanonicalAtomFeaturizer(),
edge_enc=None,
graph_type=mol_to_bigraph,
canonical_atom_order=False):
super(graph_dataset, self).__init__()
# self.num_graphs = num_graphs
self.smiles = smiles
self.y = y
self.graph_type = graph_type
self.node_enc = node_enc
self.edge_enc = edge_enc
self.canonical_atom_order = canonical_atom_order
self.graphs = []
self.labels = []
self._generate()
def main(args):
dataset = UnlabeledSMILES(args['smiles'],
node_featurizer=CanonicalAtomFeaturizer())
dataloader = DataLoader(dataset,
batch_size=args['batch_size'],
collate_fn=collate_molgraphs_unlabeled)
model = load_model(args).to(args['device'])
checkpoint = torch.load(args['train_result_path'] + '/model.pth',
map_location='cpu')
model.load_state_dict(checkpoint['model_state_dict'])
model.eval()
smiles_list = []
predictions = []
with torch.no_grad():
for batch_id, batch_data in enumerate(
tqdm(dataloader, desc="Iteration")):
batch_smiles, bg = batch_data
smiles_list.extend(batch_smiles)
batch_pred = predict(args, model, bg)
if not args['soft_classification']:
batch_pred = (batch_pred >= 0.5).float()
predictions.append(batch_pred.detach().cpu())
predictions = torch.cat(predictions, dim=0)
output_data = {'canonical_smiles': smiles_list}
if args['task_names'] is None:
args['task_names'] = [
'task_{:d}'.format(t) for t in range(1, args['n_tasks'] + 1)
]
else:
args['task_names'] = args['task_names'].split(',')
for task_id, task_name in enumerate(args['task_names']):
output_data[task_name] = predictions[:, task_id]
df = pd.DataFrame(output_data)
df.to_csv(args['inference_result_path'] + '/prediction.csv', index=False)
def main(args):
"""
:param n_trials: int specifying number of random train/test splits to use
:param test_set_size: float in range [0, 1] specifying fraction of dataset to use as test set
"""
df = pd.read_csv('data/covid_multitask_pIC50.smi')
smiles_list = df['SMILES']
y = df[['acry_class', 'chloro_class', 'rest_class', 'acry_reg', 'chloro_reg', 'rest_reg']].to_numpy()
n_tasks = y.shape[1]
class_inds = [0,1,2]
reg_inds = [3,4,5]
X = [Chem.MolFromSmiles(m) for m in smiles_list]
# Initialise featurisers
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
e_feats = bond_featurizer.feat_size('e')
n_feats = atom_featurizer.feat_size('h')
print('Number of features: ', n_feats)
X = np.array([mol_to_bigraph(m, node_featurizer=atom_featurizer, edge_featurizer=bond_featurizer) for m in X])
r2_list = []
rmse_list = []
roc_list = []
prc_list = []
for i in range(args.n_trials):
#kf = StratifiedKFold(n_splits=3, random_state=i, shuffle=True)
#split_list = kf.split(X, y)
#X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=args.test_set_size, random_state=i+5)
X_train_acry, X_test_acry, \
y_train_acry, y_test_acry = train_test_split(X[~np.isnan(y[:,0])], y[~np.isnan(y[:,0])],
stratify=y[:,0][~np.isnan(y[:,0])],
test_size=args.test_set_size, shuffle=True, random_state=i+5)
X_train_chloro, X_test_chloro, \
y_train_chloro, y_test_chloro = train_test_split(X[~np.isnan(y[:,1])], y[~np.isnan(y[:,1])],
stratify=y[:,1][~np.isnan(y[:,1])],
test_size=args.test_set_size, shuffle=True, random_state=i+5)
X_train_rest, X_test_rest, \
y_train_rest, y_test_rest = train_test_split(X[~np.isnan(y[:,2])], y[~np.isnan(y[:,2])],
stratify=y[:,2][~np.isnan(y[:,2])],
test_size=args.test_set_size, shuffle=True, random_state=i+5)
X_train = np.concatenate([X_train_acry, X_train_chloro, X_train_rest])
X_test = np.concatenate([X_test_acry, X_test_chloro, X_test_rest])
y_train = np.concatenate([y_train_acry, y_train_chloro, y_train_rest])
y_test = np.concatenate([y_test_acry, y_test_chloro, y_test_rest])
writer = SummaryWriter('runs/multitask_pIC50/run_' + str(i))
# writer = SummaryWriter('runs/multitask/run_' + str(i) + '_fold_' + str(j))
y_train = torch.Tensor(y_train)
y_test = torch.Tensor(y_test)
train_data = list(zip(X_train, y_train))
test_data = list(zip(X_test, y_test))
train_loader = DataLoader(train_data, batch_size=32, shuffle=True, collate_fn=collate, drop_last=False)
test_loader = DataLoader(test_data, batch_size=32, shuffle=True, collate_fn=collate, drop_last=False)
process = Net(class_inds, reg_inds)
process.to(device)
mpnn_net = MPNNPredictor(node_in_feats=n_feats,
edge_in_feats=e_feats,
node_out_feats=128,
n_tasks=n_tasks
)
mpnn_net.to(device)
reg_loss_fn = MSELoss()
class_loss_fn = BCELoss()
optimizer = torch.optim.Adam(mpnn_net.parameters(), lr=0.001)
for epoch in range(1, args.n_epochs+1):
epoch_loss = 0
preds = []
labs = []
mpnn_net.train()
for i, (bg, labels) in enumerate(train_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
#atom_feats, bond_feats, dcs, labels = atom_feats.to(device), bond_feats.to(device), dcs.to(device), labels.to(device)
y_pred = mpnn_net(bg, atom_feats, bond_feats)
y_pred = process(y_pred)
loss=torch.tensor(0)
loss = loss.to(device)
for ind in reg_inds:
if len(labels[:,ind][~torch.isnan(labels[:,ind])])==0:
continue
loss = loss + reg_loss_fn(y_pred[:,ind][~torch.isnan(labels[:,ind])], labels[:,ind][~torch.isnan(labels[:,ind])])
for ind in class_inds:
if len(labels[:,ind][~torch.isnan(labels[:,ind])])==0:
continue
loss = loss + class_loss_fn(y_pred[:,ind][~torch.isnan(labels[:,ind])], labels[:,ind][~torch.isnan(labels[:,ind])])
optimizer.zero_grad()
loss.backward()
optimizer.step()
epoch_loss += loss.detach().item()
labels = labels.cpu().numpy()
y_pred = y_pred.detach().cpu().numpy()
# store labels and preds
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=0)
preds = np.concatenate(preds, axis=0)
rmses= []
r2s = []
rocs = []
prcs = []
for ind in reg_inds:
rmse = np.sqrt(mean_squared_error(labs[:,ind][~np.isnan(labs[:,ind])],preds[:,ind][~np.isnan(labs[:,ind])]))
r2 = r2_score(labs[:,ind][~np.isnan(labs[:,ind])],preds[:,ind][~np.isnan(labs[:,ind])])
rmses.append(rmse)
r2s.append(r2)
for ind in class_inds:
r2 = roc_auc_score(labs[:,ind][~np.isnan(labs[:,ind])],
preds[:,ind][~np.isnan(labs[:,ind])])
precision, recall, thresholds = precision_recall_curve(labs[:,ind][~np.isnan(labs[:,ind])],
preds[:,ind][~np.isnan(labs[:,ind])])
rmse = auc(recall, precision)
rocs.append(r2)
prcs.append(rmse)
writer.add_scalar('LOSS/train', epoch_loss, epoch)
writer.add_scalar('train/acry_rocauc', rocs[0], epoch)
writer.add_scalar('train/acry_prcauc', prcs[0], epoch)
writer.add_scalar('train/chloro_rocauc', rocs[1], epoch)
writer.add_scalar('train/chloro_prcauc', prcs[1], epoch)
writer.add_scalar('train/rest_rocauc', rocs[2], epoch)
writer.add_scalar('train/rest_prcauc', prcs[2], epoch)
writer.add_scalar('train/acry_rmse', rmses[0], epoch)
writer.add_scalar('train/acry_r2', r2s[0], epoch)
writer.add_scalar('train/chloro_rmse', rmses[1], epoch)
writer.add_scalar('train/chloro_r2', r2s[1], epoch)
writer.add_scalar('train/rest_rmse', rmses[2], epoch)
writer.add_scalar('train/rest_r2', r2s[2], epoch)
if epoch % 20 == 0:
print(f"\nepoch: {epoch}, "
f"LOSS: {epoch_loss:.3f}"
f"\n acry ROC-AUC: {rocs[0]:.3f}, "
f"acry PRC-AUC: {prcs[0]:.3f}"
f"\n chloro ROC-AUC: {rocs[1]:.3f}, "
f"chloro PRC-AUC: {prcs[1]:.3f}"
f"\n rest ROC-AUC: {rocs[2]:.3f}, "
f"rest PRC-AUC: {prcs[2]:.3f}"
f"\n acry R2: {r2s[0]:.3f}, "
f"acry RMSE: {rmses[0]:.3f}"
f"\n chloro R2: {r2s[1]:.3f}, "
f"chloro RMSE: {rmses[1]:.3f}"
f"\n rest R2: {r2s[2]:.3f}, "
f"rest RMSE: {rmses[2]:.3f}")
# Evaluate
mpnn_net.eval()
preds = []
labs = []
for i, (bg, labels) in enumerate(test_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
#atom_feats, bond_feats, labels = atom_feats.to(device), bond_feats.to(device), labels.to(device)
y_pred = mpnn_net(bg, atom_feats, bond_feats)
y_pred = process(y_pred)
labels = labels.cpu().numpy()
y_pred = y_pred.detach().cpu().numpy()
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=0)
preds = np.concatenate(preds, axis=0)
rmses = []
r2s = []
rocs = []
prcs = []
for ind in reg_inds:
rmse = np.sqrt(mean_squared_error(labs[:,ind][~np.isnan(labs[:,ind])],preds[:,ind][~np.isnan(labs[:,ind])]))
r2 = r2_score(labs[:,ind][~np.isnan(labs[:,ind])],preds[:,ind][~np.isnan(labs[:,ind])])
rmses.append(rmse)
r2s.append(r2)
for ind in class_inds:
r2 = roc_auc_score(labs[:, ind][~np.isnan(labs[:,ind])],
preds[:, ind][~np.isnan(labs[:,ind])])
precision, recall, thresholds = precision_recall_curve(labs[:, ind][~np.isnan(labs[:,ind])],
preds[:, ind][~np.isnan(labs[:,ind])])
rmse = auc(recall, precision)
rocs.append(r2)
prcs.append(rmse)
writer.add_scalar('test/acry_rocauc', rocs[0], epoch)
writer.add_scalar('test/acry_prcauc', prcs[0], epoch)
writer.add_scalar('test/chloro_rocauc', rocs[1], epoch)
writer.add_scalar('test/chloro_prcauc', prcs[1], epoch)
writer.add_scalar('test/rest_rocauc', rocs[2], epoch)
writer.add_scalar('test/rest_prcauc', prcs[2], epoch)
writer.add_scalar('test/acry_rmse', rmses[0], epoch)
writer.add_scalar('test/acry_r2', r2s[0], epoch)
writer.add_scalar('test/chloro_rmse', rmses[1], epoch)
writer.add_scalar('test/chloro_r2', r2s[1], epoch)
writer.add_scalar('test/rest_rmse', rmses[2], epoch)
writer.add_scalar('test/rest_r2', r2s[2], epoch)
if epoch==(args.n_epochs):
print(f"\n======================== TEST ========================"
f"\n acry ROC-AUC: {rocs[0]:.3f}, "
f"acry PRC-AUC: {prcs[0]:.3f}"
f"\n chloro ROC-AUC: {rocs[1]:.3f}, "
f"chloro PRC-AUC: {prcs[1]:.3f}"
f"\n rest ROC-AUC: {rocs[2]:.3f}, "
f"rest PRC-AUC: {prcs[2]:.3f}"
f"\n acry R2: {r2s[0]:.3f}, "
f"acry RMSE: {rmses[0]:.3f}"
f"\n chloro R2: {r2s[1]:.3f}, "
f"chloro RMSE: {rmses[1]:.3f}"
f"\n rest R2: {r2s[2]:.3f}, "
f"rest RMSE: {rmses[2]:.3f}")
roc_list.append(rocs)
prc_list.append(prcs)
r2_list.append(r2s)
rmse_list.append(rmses)
roc_list = np.array(roc_list).T
prc_list = np.array(prc_list).T
r2_list = np.array(r2_list).T
rmse_list = np.array(rmse_list).T
print("\n ACRY")
print("R^2: {:.4f} +- {:.4f}".format(np.mean(r2_list[0]), np.std(r2_list[0])/np.sqrt(len(r2_list[0]))))
print("RMSE: {:.4f} +- {:.4f}".format(np.mean(rmse_list[0]), np.std(rmse_list[0])/np.sqrt(len(rmse_list[0]))))
print("ROC-AUC: {:.3f} +- {:.3f}".format(np.mean(roc_list[0]), np.std(roc_list[0]) / np.sqrt(len(roc_list[0]))))
print("PRC-AUC: {:.3f} +- {:.3f}".format(np.mean(prc_list[0]), np.std(prc_list[0]) / np.sqrt(len(prc_list[0]))))
print("\n CHLORO")
print("R^2: {:.4f} +- {:.4f}".format(np.mean(r2_list[1]), np.std(r2_list[1])/np.sqrt(len(r2_list[1]))))
print("RMSE: {:.4f} +- {:.4f}".format(np.mean(rmse_list[1]), np.std(rmse_list[1])/np.sqrt(len(rmse_list[1]))))
print("ROC-AUC: {:.3f} +- {:.3f}".format(np.mean(roc_list[1]), np.std(roc_list[1]) / np.sqrt(len(roc_list[1]))))
print("PRC-AUC: {:.3f} +- {:.3f}".format(np.mean(prc_list[1]), np.std(prc_list[1]) / np.sqrt(len(prc_list[1]))))
print("\n REST")
print("R^2: {:.4f} +- {:.4f}".format(np.mean(r2_list[2]), np.std(r2_list[2])/np.sqrt(len(r2_list[2]))))
print("RMSE: {:.4f} +- {:.4f}".format(np.mean(rmse_list[2]), np.std(rmse_list[2])/np.sqrt(len(rmse_list[2]))))
print("ROC-AUC: {:.3f} +- {:.3f}".format(np.mean(roc_list[2]), np.std(roc_list[2]) / np.sqrt(len(roc_list[2]))))
print("PRC-AUC: {:.3f} +- {:.3f}".format(np.mean(prc_list[2]), np.std(prc_list[2]) / np.sqrt(len(prc_list[2]))))
def main(args):
"""
:param path: str specifying path to dataset.
:param task: str specifying the task. One of ['e_iso_pi', 'z_iso_pi', 'e_iso_n', 'z_iso_n']
:param n_trials: int specifying number of random train/test splits to use
:param test_set_size: float in range [0, 1] specifying fraction of dataset to use as test set
"""
# data_loader = TaskDataLoader(args.task, args.path)
# smiles_list, y = data_loader.load_property_data()
smiles_list, y = parse_dataset(args.task, PATHS[args.task], args.reg)
X = [Chem.MolFromSmiles(m) for m in smiles_list]
# Initialise featurisers
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
e_feats = bond_featurizer.feat_size('e')
n_feats = atom_featurizer.feat_size('h')
print('Number of features: ', n_feats)
X = [
mol_to_bigraph(m,
node_featurizer=atom_featurizer,
edge_featurizer=bond_featurizer) for m in X
]
r2_list = []
rmse_list = []
mae_list = []
skipped_trials = 0
for i in range(args.n_trials):
# X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=args.test_set_size, random_state=i + 5)
kf = StratifiedKFold(n_splits=args.n_folds,
random_state=i,
shuffle=True)
split_list = kf.split(X, y)
j = 0
for train_ind, test_ind in split_list:
if args.reg:
writer = SummaryWriter('runs/' + args.task + '/mpnn/reg/run_' +
str(i) + '_fold_' + str(j))
else:
writer = SummaryWriter('runs/' + args.task +
'/mpnn/class/run_' + str(i) + '_fold_' +
str(j))
X_train, X_test = np.array(X)[train_ind], np.array(X)[test_ind]
y_train, y_test = np.array(y)[train_ind], np.array(y)[test_ind]
y_train = y_train.reshape(-1, 1)
y_test = y_test.reshape(-1, 1)
# We standardise the outputs but leave the inputs unchanged
if args.reg:
y_scaler = StandardScaler()
y_train_scaled = torch.Tensor(y_scaler.fit_transform(y_train))
y_test_scaled = torch.Tensor(y_scaler.transform(y_test))
else:
y_train_scaled = torch.Tensor(y_train)
y_test_scaled = torch.Tensor(y_test)
train_data = list(zip(X_train, y_train_scaled))
test_data = list(zip(X_test, y_test_scaled))
train_loader = DataLoader(train_data,
batch_size=32,
shuffle=True,
collate_fn=collate,
drop_last=False)
test_loader = DataLoader(test_data,
batch_size=32,
shuffle=False,
collate_fn=collate,
drop_last=False)
mpnn_net = MPNNPredictor(node_in_feats=n_feats,
edge_in_feats=e_feats)
mpnn_net.to(device)
if args.reg:
loss_fn = MSELoss()
else:
loss_fn = BCELoss()
optimizer = torch.optim.Adam(mpnn_net.parameters(), lr=1e-4)
mpnn_net.train()
epoch_losses = []
epoch_rmses = []
for epoch in tqdm(range(1, args.n_epochs)):
epoch_loss = 0
preds = []
labs = []
for i, (bg, labels) in tqdm(enumerate(train_loader)):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
atom_feats, bond_feats, labels = atom_feats.to(
device), bond_feats.to(device), labels.to(device)
y_pred = mpnn_net(bg, atom_feats, bond_feats)
labels = labels.unsqueeze(dim=1)
loss = loss_fn(y_pred, labels)
optimizer.zero_grad()
loss.backward()
optimizer.step()
epoch_loss += loss.detach().item()
if args.reg:
# Inverse transform to get RMSE
labels = y_scaler.inverse_transform(
labels.cpu().reshape(-1, 1))
y_pred = y_scaler.inverse_transform(
y_pred.detach().cpu().numpy().reshape(-1, 1))
else:
labels = labels.cpu().numpy()
y_pred = y_pred.detach().cpu().numpy()
# store labels and preds
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=None)
preds = np.concatenate(preds, axis=None)
pearson, p = pearsonr(preds, labs)
if args.reg:
mae = mean_absolute_error(preds, labs)
rmse = np.sqrt(mean_squared_error(preds, labs))
r2 = r2_score(preds, labs)
else:
r2 = roc_auc_score(labs, preds)
precision, recall, thresholds = precision_recall_curve(
labs, preds)
rmse = auc(recall, precision)
mae = 0
if args.reg:
writer.add_scalar('Loss/train', epoch_loss, epoch)
writer.add_scalar('RMSE/train', rmse, epoch)
writer.add_scalar('R2/train', r2, epoch)
else:
writer.add_scalar('Loss/train', epoch_loss, epoch)
writer.add_scalar('ROC-AUC/train', r2, epoch)
writer.add_scalar('PRC-AUC/train', rmse, epoch)
if epoch % 20 == 0:
if args.reg:
print(f"epoch: {epoch}, "
f"LOSS: {epoch_loss:.3f}, "
f"RMSE: {rmse:.3f}, "
f"MAE: {mae:.3f}, "
f"rho: {pearson:.3f}, "
f"R2: {r2:.3f}")
else:
print(f"epoch: {epoch}, "
f"LOSS: {epoch_loss:.3f}, "
f"ROC-AUC: {r2:.3f}, "
f"PRC-AUC: {rmse:.3f}, "
f"rho: {pearson:.3f}")
epoch_losses.append(epoch_loss)
epoch_rmses.append(rmse)
# Discount trial if train RMSE finishes as a negative value (optimiser error).
if r2 < -1:
skipped_trials += 1
print('Skipped trials is {}'.format(skipped_trials))
continue
# Evaluate
mpnn_net.eval()
preds = []
labs = []
for i, (bg, labels) in enumerate(test_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
atom_feats, bond_feats, labels = atom_feats.to(
device), bond_feats.to(device), labels.to(device)
y_pred = mpnn_net(bg, atom_feats, bond_feats)
labels = labels.unsqueeze(dim=1)
if args.reg:
# Inverse transform to get RMSE
labels = y_scaler.inverse_transform(labels.cpu().reshape(
-1, 1))
y_pred = y_scaler.inverse_transform(
y_pred.detach().cpu().numpy().reshape(-1, 1))
else:
labels = labels.cpu().numpy()
y_pred = y_pred.detach().cpu().numpy()
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=None)
preds = np.concatenate(preds, axis=None)
pearson, p = pearsonr(preds, labs)
if args.reg:
mae = mean_absolute_error(preds, labs)
rmse = np.sqrt(mean_squared_error(preds, labs))
r2 = r2_score(preds, labs)
writer.add_scalar('RMSE/test', rmse)
writer.add_scalar('R2/test', r2)
print(
f'Test RMSE: {rmse:.3f}, MAE: {mae:.3f}, R: {pearson:.3f}, R2: {r2:.3f}'
)
else:
r2 = roc_auc_score(labs, preds)
precision, recall, thresholds = precision_recall_curve(
labs, preds)
rmse = auc(recall, precision)
mae = 0
writer.add_scalar('ROC-AUC/test', r2)
writer.add_scalar('PRC-AUC/test', rmse)
print(
f'Test ROC-AUC: {r2:.3f}, PRC-AUC: {rmse:.3f}, rho: {pearson:.3f}'
)
r2_list.append(r2)
rmse_list.append(rmse)
mae_list.append(mae)
j += 1
r2_list = np.array(r2_list)
rmse_list = np.array(rmse_list)
mae_list = np.array(mae_list)
if args.reg:
print("\nmean R^2: {:.4f} +- {:.4f}".format(
np.mean(r2_list),
np.std(r2_list) / np.sqrt(len(r2_list))))
print("mean RMSE: {:.4f} +- {:.4f}".format(
np.mean(rmse_list),
np.std(rmse_list) / np.sqrt(len(rmse_list))))
print("mean MAE: {:.4f} +- {:.4f}\n".format(
np.mean(mae_list),
np.std(mae_list) / np.sqrt(len(mae_list))))
else:
print("mean ROC-AUC^2: {:.3f} +- {:.3f}".format(
np.mean(r2_list),
np.std(r2_list) / np.sqrt(len(r2_list))))
print("mean PRC-AUC: {:.3f} +- {:.3f}".format(
np.mean(rmse_list),
np.std(rmse_list) / np.sqrt(len(rmse_list))))
print("\nSkipped trials is {}".format(skipped_trials))
test_y = np.array(test_y, dtype=np.int64)
print("Test set", len(test_g))
print("Data loaded.")
return train_g, train_y, test_g, test_y
def collate(sample):
graphs, labels = map(list, zip(*sample))
batched_graph = dgl.batch(graphs)
batched_graph.set_n_initializer(dgl.init.zero_initializer)
batched_graph.set_e_initializer(dgl.init.zero_initializer)
return batched_graph, torch.tensor(labels)
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
def featurize_edges(mol, add_self_loop=False):
feats = []
num_atoms = mol.GetNumAtoms()
atoms = list(mol.GetAtoms())
distance_matrix = Chem.GetDistanceMatrix(mol)
for i in range(num_atoms):
for j in range(num_atoms):
if i != j or add_self_loop:
feats.append(float(distance_matrix[i, j]))
return {'e': torch.tensor(feats).reshape(-1, 1).float()}
def _get_collection(collection):
from dgllife.utils import smiles_to_bigraph, CanonicalAtomFeaturizer
ds = getattr(dgllife.data, collection)(smiles_to_bigraph,
CanonicalAtomFeaturizer())
return _dataset_from_dgllife(ds)
help='Print the training progress every X mini-batches')
parser.add_argument('-p', '--result-path', type=str, default='classification_results',
help='Path to save training results (default: classification_results)')
parser.add_argument('-ne', '--num-evals', type=int, default=None,
help='Number of trials for hyperparameter search (default: None)')
args = parser.parse_args().__dict__
if torch.cuda.is_available():
args['device'] = torch.device('cuda:0')
else:
args['device'] = torch.device('cpu')
if args['task_names'] is not None:
args['task_names'] = args['task_names'].split(',')
args['node_featurizer'] = CanonicalAtomFeaturizer()
df = pd.read_csv(args['csv_path'])
mkdir_p(args['result_path'])
dataset = MoleculeCSVDataset(df=df,
smiles_to_graph=smiles_to_bigraph,
node_featurizer=args['node_featurizer'],
edge_featurizer=None,
smiles_column=args['smiles_column'],
cache_file_path=args['result_path'] + '/graph.bin',
task_names=args['task_names'])
args['n_tasks'] = dataset.n_tasks
train_set, val_set, test_set = split_dataset(args, dataset)
if args['num_evals'] is not None:
assert args['num_evals'] > 0, 'Expect the number of hyperparameter search trials to ' \
'be greater than 0, got {:d}'.format(args['num_evals'])
def main(path, task, n_trials, test_set_size):
"""
:param path: str specifying path to dataset.
:param task: str specifying the task. One of ['e_iso_pi', 'z_iso_pi', 'e_iso_n', 'z_iso_n']
:param n_trials: int specifying number of random train/test splits to use
:param test_set_size: float in range [0, 1] specifying fraction of dataset to use as test set.
"""
data_loader = TaskDataLoader(task, path)
smiles_list, y = data_loader.load_property_data()
X = [Chem.MolFromSmiles(m) for m in smiles_list]
# Collate Function for Dataloader
def collate(sample):
graphs, labels = map(list, zip(*sample))
batched_graph = dgl.batch(graphs)
batched_graph.set_n_initializer(dgl.init.zero_initializer)
batched_graph.set_e_initializer(dgl.init.zero_initializer)
return batched_graph, torch.tensor(labels)
# Initialise featurisers
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
e_feats = bond_featurizer.feat_size('e')
n_feats = atom_featurizer.feat_size('h')
print('Number of features: ', n_feats)
X = [
mol_to_bigraph(m,
node_featurizer=atom_featurizer,
edge_featurizer=bond_featurizer) for m in X
]
r2_list = []
rmse_list = []
mae_list = []
skipped_trials = 0
for i in range(0, n_trials):
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size=test_set_size, random_state=i + 5)
y_train = y_train.reshape(-1, 1)
y_test = y_test.reshape(-1, 1)
# We standardise the outputs but leave the inputs unchanged
y_scaler = StandardScaler()
y_train_scaled = torch.Tensor(y_scaler.fit_transform(y_train))
y_test_scaled = torch.Tensor(y_scaler.transform(y_test))
train_data = list(zip(X_train, y_train_scaled))
test_data = list(zip(X_test, y_test_scaled))
train_loader = DataLoader(train_data,
batch_size=32,
shuffle=True,
collate_fn=collate,
drop_last=False)
test_loader = DataLoader(test_data,
batch_size=32,
shuffle=False,
collate_fn=collate,
drop_last=False)
gat_net = GATPredictor(in_feats=n_feats)
gat_net.to(device)
loss_fn = MSELoss()
optimizer = torch.optim.Adam(gat_net.parameters(), lr=0.001)
gat_net.train()
epoch_losses = []
epoch_rmses = []
for epoch in range(1, 201):
epoch_loss = 0
preds = []
labs = []
for i, (bg, labels) in enumerate(train_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
atom_feats, bond_feats, labels = atom_feats.to(
device), bond_feats.to(device), labels.to(device)
y_pred = gat_net(bg, atom_feats)
labels = labels.unsqueeze(dim=1)
loss = loss_fn(y_pred, labels)
optimizer.zero_grad()
loss.backward()
optimizer.step()
epoch_loss += loss.detach().item()
# Inverse transform to get RMSE
labels = y_scaler.inverse_transform(labels.reshape(-1, 1))
y_pred = y_scaler.inverse_transform(
y_pred.detach().numpy().reshape(-1, 1))
# store labels and preds
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=None)
preds = np.concatenate(preds, axis=None)
pearson, p = pearsonr(preds, labs)
mae = mean_absolute_error(preds, labs)
rmse = np.sqrt(mean_squared_error(preds, labs))
r2 = r2_score(preds, labs)
epoch_loss /= (i + 1)
if epoch % 20 == 0:
print(f"epoch: {epoch}, "
f"LOSS: {epoch_loss:.3f}, "
f"RMSE: {rmse:.3f}, "
f"MAE: {mae:.3f}, "
f"R: {pearson:.3f}, "
f"R2: {r2:.3f}")
epoch_losses.append(epoch_loss)
epoch_rmses.append(rmse)
# Discount trial if train RMSE finishes as a negative value (optimiser error).
if r2 < 0:
skipped_trials += 1
print('Skipped trials is {}'.format(skipped_trials))
continue
# Evaluate
gat_net.eval()
preds = []
labs = []
for i, (bg, labels) in enumerate(test_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
atom_feats, labels = atom_feats.to(device), labels.to(device)
y_pred = gat_net(bg, atom_feats)
labels = labels.unsqueeze(dim=1)
# Inverse transform to get RMSE
labels = y_scaler.inverse_transform(labels.reshape(-1, 1))
y_pred = y_scaler.inverse_transform(
y_pred.detach().numpy().reshape(-1, 1))
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=None)
preds = np.concatenate(preds, axis=None)
pearson, p = pearsonr(preds, labs)
mae = mean_absolute_error(preds, labs)
rmse = np.sqrt(mean_squared_error(preds, labs))
r2 = r2_score(preds, labs)
r2_list.append(r2)
rmse_list.append(rmse)
mae_list.append(mae)
print(
f'Test RMSE: {rmse:.3f}, MAE: {mae:.3f}, R: {pearson:.3f}, R2: {r2:.3f}'
)
r2_list = np.array(r2_list)
rmse_list = np.array(rmse_list)
mae_list = np.array(mae_list)
print("\nmean R^2: {:.4f} +- {:.4f}".format(
np.mean(r2_list),
np.std(r2_list) / np.sqrt(len(r2_list))))
print("mean RMSE: {:.4f} +- {:.4f}".format(
np.mean(rmse_list),
np.std(rmse_list) / np.sqrt(len(rmse_list))))
print("mean MAE: {:.4f} +- {:.4f}\n".format(
np.mean(mae_list),
np.std(mae_list) / np.sqrt(len(mae_list))))
print("\nSkipped trials is {}".format(skipped_trials))
def name2g(data_path, name):
path = os.path.join(data_path, name + '.sdf')
for mol in Chem.SDMolSupplier(path):
g = mol_to_bigraph(mol, node_featurizer=CanonicalAtomFeaturizer())
return g
def main(args):
"""
:param n_trials: int specifying number of random train/test splits to use
:param test_set_size: float in range [0, 1] specifying fraction of dataset to use as test set
"""
df = pd.read_csv('data/covid_multitask_HTS.smi')
if args.dry:
df = df[:2000]
smiles_list = df['SMILES'].values
y = df[[
'acry_reg', 'chloro_reg', 'rest_reg', 'acry_class', 'chloro_class',
'rest_class', 'activity'
]].to_numpy()
n_tasks = y.shape[1]
reg_inds = [0, 1, 2]
class_inds = [3, 4, 5, 6]
# print(smiles_list)
X = [Chem.MolFromSmiles(m) for m in smiles_list]
# Initialise featurisers
atom_featurizer = CanonicalAtomFeaturizer()
bond_featurizer = CanonicalBondFeaturizer()
e_feats = bond_featurizer.feat_size('e')
n_feats = atom_featurizer.feat_size('h')
print('Number of features: ', n_feats)
X = np.array([
mol_to_bigraph(m,
node_featurizer=atom_featurizer,
edge_featurizer=bond_featurizer) for m in X
])
r2_list = []
rmse_list = []
roc_list = []
prc_list = []
for i in range(args.n_trials):
writer = SummaryWriter('runs/' + args.savename)
if args.test:
X_train_acry, X_test_acry, \
y_train_acry, y_test_acry = train_test_split(X[~np.isnan(y[:,3])],
y[~np.isnan(y[:,3])], stratify=y[:,3][~np.isnan(y[:,3])],
test_size=args.test_set_size, shuffle=True, random_state=i+5)
X_train_chloro, X_test_chloro, \
y_train_chloro, y_test_chloro = train_test_split(X[~np.isnan(y[:,4])],
y[~np.isnan(y[:,4])], stratify=y[:,4][~np.isnan(y[:,4])],
test_size=args.test_set_size, shuffle=True, random_state=i+5)
X_train_rest, X_test_rest, \
y_train_rest, y_test_rest = train_test_split(X[~np.isnan(y[:,5])],
y[~np.isnan(y[:,5])], stratify=y[:,5][~np.isnan(y[:,5])],
test_size=args.test_set_size, shuffle=True, random_state=i+5)
X_train = np.concatenate([
X_train_acry, X_train_chloro, X_train_rest,
X[~np.isnan(y[:, 6])]
])
X_test = np.concatenate([X_test_acry, X_test_chloro, X_test_rest])
y_train = np.concatenate([
y_train_acry, y_train_chloro, y_train_rest,
y[~np.isnan(y[:, 6])]
])
y_test = np.concatenate([y_test_acry, y_test_chloro, y_test_rest])
y_train = torch.Tensor(y_train)
y_test = torch.Tensor(y_test)
train_data = list(zip(X_train, y_train))
test_data = list(zip(X_test, y_test))
train_loader = DataLoader(train_data,
batch_size=32,
shuffle=True,
collate_fn=collate,
drop_last=False)
test_loader = DataLoader(test_data,
batch_size=32,
shuffle=True,
collate_fn=collate,
drop_last=False)
else:
y = torch.Tensor(y)
train_data = list(zip(X, y))
train_loader = DataLoader(train_data,
batch_size=32,
shuffle=True,
collate_fn=collate,
drop_last=False)
process = Net(class_inds, reg_inds)
process = process.to(device)
mpnn_net = MPNNPredictor(node_in_feats=n_feats,
edge_in_feats=e_feats,
node_out_feats=128,
n_tasks=n_tasks)
mpnn_net = mpnn_net.to(device)
# try:
# mpnn_net.load_state_dict(torch.load('/rds-d2/user/wjm41/hpc-work/models/' + args.savename +
# '/model_epoch_20.pt'))
# except: pass
reg_loss_fn = MSELoss()
class_loss_fn = BCELoss()
optimizer = torch.optim.Adam(mpnn_net.parameters(), lr=1e-4)
for epoch in range(1, args.n_epochs + 1):
epoch_loss = 0
preds = []
labs = []
mpnn_net.train()
n = 0
for i, (bg, labels) in enumerate(train_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
y_pred = mpnn_net(bg, atom_feats, bond_feats)
y_pred = process(y_pred)
loss = torch.tensor(0)
loss = loss.to(device)
if args.debug:
print('label: {}'.format(labels))
print('y_pred: {}'.format(y_pred))
for ind in reg_inds:
if len(labels[:, ind][~torch.isnan(labels[:, ind])]) == 0:
continue
loss = loss + reg_loss_fn(
y_pred[:, ind][~torch.isnan(labels[:, ind])],
labels[:, ind][~torch.isnan(labels[:, ind])])
if args.debug:
print('reg loss: {}'.format(loss))
for ind in class_inds:
if len(labels[:, ind][~torch.isnan(labels[:, ind])]) == 0:
continue
loss = loss + class_loss_fn(
y_pred[:, ind][~torch.isnan(labels[:, ind])],
labels[:, ind][~torch.isnan(labels[:, ind])])
if args.debug:
print('class + reg loss: {}'.format(loss))
optimizer.zero_grad()
loss.backward()
optimizer.step()
epoch_loss += loss.detach().item()
labels = labels.cpu().numpy()
y_pred = y_pred.detach().cpu().numpy()
# store labels and preds
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=0)
preds = np.concatenate(preds, axis=0)
rmses = []
r2s = []
rocs = []
prcs = []
for ind in reg_inds:
rmse = np.sqrt(
mean_squared_error(labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])]))
r2 = r2_score(labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])])
rmses.append(rmse)
r2s.append(r2)
for ind in class_inds:
roc = roc_auc_score(labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])])
precision, recall, thresholds = precision_recall_curve(
labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])])
prc = auc(recall, precision)
rocs.append(roc)
prcs.append(prc)
writer.add_scalar('LOSS/train', epoch_loss, epoch)
writer.add_scalar('train/acry_rocauc', rocs[0], epoch)
writer.add_scalar('train/acry_prcauc', prcs[0], epoch)
writer.add_scalar('train/chloro_rocauc', rocs[1], epoch)
writer.add_scalar('train/chloro_prcauc', prcs[1], epoch)
writer.add_scalar('train/rest_rocauc', rocs[2], epoch)
writer.add_scalar('train/rest_prcauc', prcs[2], epoch)
writer.add_scalar('train/HTS_rocauc', rocs[3], epoch)
writer.add_scalar('train/HTS_prcauc', prcs[3], epoch)
writer.add_scalar('train/acry_rmse', rmses[0], epoch)
writer.add_scalar('train/acry_r2', r2s[0], epoch)
writer.add_scalar('train/chloro_rmse', rmses[1], epoch)
writer.add_scalar('train/chloro_r2', r2s[1], epoch)
writer.add_scalar('train/rest_rmse', rmses[2], epoch)
writer.add_scalar('train/rest_r2', r2s[2], epoch)
if epoch % 20 == 0:
print(f"\nepoch: {epoch}, "
f"LOSS: {epoch_loss:.3f}"
f"\n acry ROC-AUC: {rocs[0]:.3f}, "
f"acry PRC-AUC: {prcs[0]:.3f}"
f"\n chloro ROC-AUC: {rocs[1]:.3f}, "
f"chloro PRC-AUC: {prcs[1]:.3f}"
f"\n rest ROC-AUC: {rocs[2]:.3f}, "
f"rest PRC-AUC: {prcs[2]:.3f}"
f"\n HTS ROC-AUC: {rocs[3]:.3f}, "
f"HTS PRC-AUC: {prcs[3]:.3f}"
f"\n acry R2: {r2s[0]:.3f}, "
f"acry RMSE: {rmses[0]:.3f}"
f"\n chloro R2: {r2s[1]:.3f}, "
f"chloro RMSE: {rmses[1]:.3f}"
f"\n rest R2: {r2s[2]:.3f}, "
f"rest RMSE: {rmses[2]:.3f}")
try:
torch.save(
mpnn_net.state_dict(),
'/rds-d2/user/wjm41/hpc-work/models/' + args.savename +
'/model_epoch_' + str(epoch) + '.pt')
except FileNotFoundError:
cmd = 'mkdir /rds-d2/user/wjm41/hpc-work/models/' + args.savename
os.system(cmd)
torch.save(
mpnn_net.state_dict(),
'/rds-d2/user/wjm41/hpc-work/models/' + args.savename +
'/model_epoch_' + str(epoch) + '.pt')
if args.test:
# Evaluate
mpnn_net.eval()
preds = []
labs = []
for i, (bg, labels) in enumerate(test_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
bond_feats = bg.edata.pop('e').to(device)
y_pred = mpnn_net(bg, atom_feats, bond_feats)
y_pred = process(y_pred)
labels = labels.cpu().numpy()
y_pred = y_pred.detach().cpu().numpy()
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=0)
preds = np.concatenate(preds, axis=0)
rmses = []
r2s = []
rocs = []
prcs = []
for ind in reg_inds:
rmse = np.sqrt(
mean_squared_error(
labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])]))
r2 = r2_score(labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])])
rmses.append(rmse)
r2s.append(r2)
for ind in class_inds[:3]:
roc = roc_auc_score(labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])])
precision, recall, thresholds = precision_recall_curve(
labs[:, ind][~np.isnan(labs[:, ind])],
preds[:, ind][~np.isnan(labs[:, ind])])
prc = auc(recall, precision)
rocs.append(roc)
prcs.append(prc)
writer.add_scalar('test/acry_rocauc', rocs[0], epoch)
writer.add_scalar('test/acry_prcauc', prcs[0], epoch)
writer.add_scalar('test/chloro_rocauc', rocs[1], epoch)
writer.add_scalar('test/chloro_prcauc', prcs[1], epoch)
writer.add_scalar('test/rest_rocauc', rocs[2], epoch)
writer.add_scalar('test/rest_prcauc', prcs[2], epoch)
writer.add_scalar('test/acry_rmse', rmses[0], epoch)
writer.add_scalar('test/acry_r2', r2s[0], epoch)
writer.add_scalar('test/chloro_rmse', rmses[1], epoch)
writer.add_scalar('test/chloro_r2', r2s[1], epoch)
writer.add_scalar('test/rest_rmse', rmses[2], epoch)
writer.add_scalar('test/rest_r2', r2s[2], epoch)
if epoch == (args.n_epochs):
print(
f"\n======================== TEST ========================"
f"\n acry ROC-AUC: {rocs[0]:.3f}, "
f"acry PRC-AUC: {prcs[0]:.3f}"
f"\n chloro ROC-AUC: {rocs[1]:.3f}, "
f"chloro PRC-AUC: {prcs[1]:.3f}"
f"\n rest ROC-AUC: {rocs[2]:.3f}, "
f"rest PRC-AUC: {prcs[2]:.3f}"
f"\n acry R2: {r2s[0]:.3f}, "
f"acry RMSE: {rmses[0]:.3f}"
f"\n chloro R2: {r2s[1]:.3f}, "
f"chloro RMSE: {rmses[1]:.3f}"
f"\n rest R2: {r2s[2]:.3f}, "
f"rest RMSE: {rmses[2]:.3f}")
roc_list.append(rocs)
prc_list.append(prcs)
r2_list.append(r2s)
rmse_list.append(rmses)
torch.save(
mpnn_net.state_dict(), '/rds-d2/user/wjm41/hpc-work/models/' +
args.savename + '/model_epoch_final.pt')
if args.test:
roc_list = np.array(roc_list).T
prc_list = np.array(prc_list).T
r2_list = np.array(r2_list).T
rmse_list = np.array(rmse_list).T
print("\n ACRY")
print("R^2: {:.4f} +- {:.4f}".format(
np.mean(r2_list[0]),
np.std(r2_list[0]) / np.sqrt(len(r2_list[0]))))
print("RMSE: {:.4f} +- {:.4f}".format(
np.mean(rmse_list[0]),
np.std(rmse_list[0]) / np.sqrt(len(rmse_list[0]))))
print("ROC-AUC: {:.3f} +- {:.3f}".format(
np.mean(roc_list[0]),
np.std(roc_list[0]) / np.sqrt(len(roc_list[0]))))
print("PRC-AUC: {:.3f} +- {:.3f}".format(
np.mean(prc_list[0]),
np.std(prc_list[0]) / np.sqrt(len(prc_list[0]))))
print("\n CHLORO")
print("R^2: {:.4f} +- {:.4f}".format(
np.mean(r2_list[1]),
np.std(r2_list[1]) / np.sqrt(len(r2_list[1]))))
print("RMSE: {:.4f} +- {:.4f}".format(
np.mean(rmse_list[1]),
np.std(rmse_list[1]) / np.sqrt(len(rmse_list[1]))))
print("ROC-AUC: {:.3f} +- {:.3f}".format(
np.mean(roc_list[1]),
np.std(roc_list[1]) / np.sqrt(len(roc_list[1]))))
print("PRC-AUC: {:.3f} +- {:.3f}".format(
np.mean(prc_list[1]),
np.std(prc_list[1]) / np.sqrt(len(prc_list[1]))))
print("\n REST")
print("R^2: {:.4f} +- {:.4f}".format(
np.mean(r2_list[2]),
np.std(r2_list[2]) / np.sqrt(len(r2_list[2]))))
print("RMSE: {:.4f} +- {:.4f}".format(
np.mean(rmse_list[2]),
np.std(rmse_list[2]) / np.sqrt(len(rmse_list[2]))))
print("ROC-AUC: {:.3f} +- {:.3f}".format(
np.mean(roc_list[2]),
np.std(roc_list[2]) / np.sqrt(len(roc_list[2]))))
print("PRC-AUC: {:.3f} +- {:.3f}".format(
np.mean(prc_list[2]),
np.std(prc_list[2]) / np.sqrt(len(prc_list[2]))))
def main(path, task, graph_type):
"""
:param path: str specifying path to dataset.
:param task: str specifying the task. One of ['e_iso_pi', 'z_iso_pi', 'e_iso_n', 'z_iso_n']
:param graph_type: str. either 'bigraph' or 'complete'
"""
data_loader = TaskDataLoader(task, path)
X, y = data_loader.load_property_data()
X = [Chem.MolFromSmiles(m) for m in X]
# Collate Function for Dataloader
def collate(sample):
graphs, labels = map(list, zip(*sample))
batched_graph = dgl.batch(graphs)
batched_graph.set_n_initializer(dgl.init.zero_initializer)
batched_graph.set_e_initializer(dgl.init.zero_initializer)
return batched_graph, torch.tensor(labels)
# Initialise featurisers
atom_featurizer = CanonicalAtomFeaturizer()
n_feats = atom_featurizer.feat_size('h')
print('Number of features: ', n_feats)
X_full, _, y_full, _ = train_test_split(X,
y,
test_size=0.2,
random_state=30)
y_full = y_full.reshape(-1, 1)
# We standardise the outputs but leave the inputs unchanged
y_scaler = StandardScaler()
y_full = torch.Tensor(y_scaler.fit_transform(y_full))
# Set up cross-validation splits
n_splits = 5
kf = KFold(n_splits=n_splits)
X_train_splits = []
y_train_splits = []
X_val_splits = []
y_val_splits = []
for train_index, test_index in kf.split(X_full):
X_train, X_val = np.array(X_full)[train_index], np.array(
X_full)[test_index]
y_train, y_val = y_full[train_index], y_full[test_index]
# Create graphs and labels
if graph_type == 'complete':
X_train = [
mol_to_complete_graph(m, node_featurizer=atom_featurizer)
for m in X_train
]
X_val = [
mol_to_complete_graph(m, node_featurizer=atom_featurizer)
for m in X_val
]
elif graph_type == 'bigraph':
X_train = [
mol_to_bigraph(m, node_featurizer=atom_featurizer) for m in X
]
X_val = [
mol_to_bigraph(m, node_featurizer=atom_featurizer)
for m in X_val
]
X_train_splits.append(X_train)
X_val_splits.append(X_val)
y_train_splits.append(y_train)
y_val_splits.append(y_val)
def lognuniform(low=1, high=5, size=None, base=10):
return np.power(base, -np.random.uniform(low, high, size))
best_rmse = 100000000
for i in range(1000):
num_layers = np.random.randint(1, 4)
classifier_hidden_feats = np.random.randint(1, 128)
hidden_feats = [np.random.choice([16, 32, 64])] * num_layers
dropout = [np.random.uniform(0, 0.5)] * num_layers
batchnorm = [np.random.choice([True, False])] * num_layers
learning_rate = lognuniform()
param_set = {
'num_layers': num_layers,
'classifier_hidden_feats': classifier_hidden_feats,
'hidden_feats': hidden_feats,
'dropout': dropout,
'batchnorm': batchnorm,
'lr': learning_rate
}
print(f'\nParameter set in trial {i} is \n')
print(param_set)
print('\n')
cv_rmse_list = []
for j in range(n_splits):
X_train = X_train_splits[j]
y_train = y_train_splits[j]
X_val = X_val_splits[j]
y_val = y_val_splits[j]
train_data = list(zip(X_train, y_train))
test_data = list(zip(X_val, y_val))
train_loader = DataLoader(train_data,
batch_size=32,
shuffle=True,
collate_fn=collate,
drop_last=False)
test_loader = DataLoader(test_data,
batch_size=32,
shuffle=False,
collate_fn=collate,
drop_last=False)
gcn_net = GCNPredictor(
in_feats=n_feats,
hidden_feats=hidden_feats,
batchnorm=batchnorm,
dropout=dropout,
classifier_hidden_feats=classifier_hidden_feats,
)
gcn_net.to(device)
loss_fn = MSELoss()
optimizer = torch.optim.Adam(gcn_net.parameters(),
lr=learning_rate)
gcn_net.train()
epoch_losses = []
epoch_rmses = []
for epoch in range(1, 501):
epoch_loss = 0
preds = []
labs = []
for i, (bg, labels) in enumerate(train_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
atom_feats, labels = atom_feats.to(device), labels.to(
device)
y_pred = gcn_net(bg, atom_feats)
labels = labels.unsqueeze(dim=1)
loss = loss_fn(y_pred, labels)
optimizer.zero_grad()
loss.backward()
optimizer.step()
epoch_loss += loss.detach().item()
# Inverse transform to get RMSE
labels = y_scaler.inverse_transform(labels.reshape(-1, 1))
y_pred = y_scaler.inverse_transform(
y_pred.detach().numpy().reshape(-1, 1))
# store labels and preds
preds.append(y_pred)
labs.append(labels)
labs = np.concatenate(labs, axis=None)
preds = np.concatenate(preds, axis=None)
pearson, p = pearsonr(preds, labs)
mae = mean_absolute_error(preds, labs)
rmse = np.sqrt(mean_squared_error(preds, labs))
r2 = r2_score(preds, labs)
epoch_loss /= (i + 1)
if epoch % 20 == 0:
print(f"epoch: {epoch}, "
f"LOSS: {epoch_loss:.3f}, "
f"RMSE: {rmse:.3f}, "
f"MAE: {mae:.3f}, "
f"R: {pearson:.3f}, "
f"R2: {r2:.3f}")
epoch_losses.append(epoch_loss)
epoch_rmses.append(rmse)
# Evaluate
gcn_net.eval()
preds = []
labs = []
for i, (bg, labels) in enumerate(test_loader):
labels = labels.to(device)
atom_feats = bg.ndata.pop('h').to(device)
atom_feats, labels = atom_feats.to(device), labels.to(device)
y_pred = gcn_net(bg, atom_feats)
labels = labels.unsqueeze(dim=1)
# Inverse transform to get RMSE
labels = y_scaler.inverse_transform(labels.reshape(-1, 1))
y_pred = y_scaler.inverse_transform(
y_pred.detach().numpy().reshape(-1, 1))
preds.append(y_pred)
labs.append(labels)
preds = np.concatenate(preds, axis=None)
labs = np.concatenate(labs, axis=None)
pearson, p = pearsonr(preds, labs)
mae = mean_absolute_error(preds, labs)
rmse = np.sqrt(mean_squared_error(preds, labs))
cv_rmse_list.append(rmse)
r2 = r2_score(preds, labs)
print(
f'Test RMSE: {rmse:.3f}, MAE: {mae:.3f}, R: {pearson:.3f}, R2: {r2:.3f}'
)
param_rmse = np.mean(cv_rmse_list)
if param_rmse < best_rmse:
best_rmse = param_rmse
best_params = param_set
print('Best RMSE and best params \n')
print(best_rmse)
print(best_params)
np.savetxt('saved_hypers/GCN', best_params)
df_train = pd.read_csv('reddb_reaction_train.csv')
#df =pd.read_csv('reddb-smiles.csv')
train = df_train.head(1000)
train_smiles = train['reactantSmiles']
train_y = torch.tensor(train['reactionEnergy']).reshape(-1,1).float()
if torch.cuda.is_available():
print('use GPU')
device='cuda'
else:
print('use CPU')
device='cpu'
train_mols = [Chem.MolFromSmiles(s) for s in train_smiles ]
atom_featurizer = CanonicalAtomFeaturizer(atom_data_field = 'h')
n_feats = atom_featurizer.feat_size('h')
bond_featurizer = CanonicalBondFeaturizer(bond_data_field='h')
b_feat = bond_featurizer.feat_size('h')
train_graph =[mol_to_bigraph(mol,node_featurizer=atom_featurizer,
edge_featurizer=bond_featurizer) for mol in train_mols]
ncls = 1
model = GCNPredictor(in_feats=n_feats,
hidden_feats=[60,20],
n_tasks=ncls,
predictor_dropout=0.2)
#model = AttentiveFPGNN(n_feats,b_feat,2,200)
model = model.to(device)
GCN_Tox21 = {
'random_seed': 2,
'batch_size': 128,
'lr': 1e-3,
'num_epochs': 100,
'node_data_field': 'h',
'frac_train': 0.8,
'frac_val': 0.1,
'frac_test': 0.1,
'in_feats': 74,
'gcn_hidden_feats': [64, 64],
'classifier_hidden_feats': 64,
'patience': 10,
'smiles_to_graph': smiles_to_bigraph,
'node_featurizer': CanonicalAtomFeaturizer(),
'metric_name': 'roc_auc_score'
}
GAT_Tox21 = {
'random_seed': 2,
'batch_size': 128,
'lr': 1e-3,
'num_epochs': 100,
'node_data_field': 'h',
'frac_train': 0.8,
'frac_val': 0.1,
'frac_test': 0.1,
'in_feats': 74,
'gat_hidden_feats': [32, 32],
'classifier_hidden_feats': 64,
from dgllife.utils import CanonicalAtomFeaturizer, CanonicalBondFeaturizer, smiles_to_bigraph
from data.util import is_valid_smiles
_atom_featurizer = CanonicalAtomFeaturizer(atom_data_field="x")
_bond_featurizer = CanonicalBondFeaturizer(bond_data_field="w")
class MoleculeRepresentations(object):
def __init__(self, smi, graph):
self.smi = smi
self.graph = graph
def smi2mol_repr(smi):
if not is_valid_smiles(smi):
return None
graph = smiles_to_bigraph(smi,
node_featurizer=_atom_featurizer,
edge_featurizer=_bond_featurizer)
return MoleculeRepresentations(smi, graph)
def smi2molgraph(smi):
graph = smiles_to_bigraph(smi,
node_featurizer=_atom_featurizer,
edge_featurizer=_bond_featurizer)
return graph
