def __init__(self,
radius: float = 8.0,
max_neighbors: float = 8,
step: float = 0.2):
"""
Parameters
----------
radius: float (default 8.0)
Radius of sphere for finding neighbors of atoms in unit cell.
max_neighbors: int (default 8)
Maximum number of neighbors to consider when constructing graph.
step: float (default 0.2)
Step size for Gaussian filter. This value is used when building edge features.
"""
self.radius = radius
self.max_neighbors = int(max_neighbors)
self.step = step
# load atom_init.json
data_dir = get_data_dir()
download_url(ATOM_INIT_JSON_URL, data_dir)
atom_init_json_path = os.path.join(data_dir, 'atom_init.json')
with open(atom_init_json_path, 'r') as f:
atom_init_json = json.load(f)
self.atom_features = {
int(key): np.array(value, dtype=np.float32)
for key, value in atom_init_json.items()
}
self.valid_atom_number = set(self.atom_features.keys())
def setUp(self):
"""Downloads dataset."""
download_url(
"http://deepchem.io.s3-website-us-west-1.amazonaws.com/featurized_datasets/core_grid.json"
)
json_fname = os.path.join(get_data_dir(), 'core_grid.json')
self.core_dataset = dc.data.NumpyDataset.from_json(json_fname)
def setUp(self):
"""Downloads dataset."""
download_url(
"http://deepchem.io.s3-website-us-west-1.amazonaws.com/featurized_datasets/core_grid.json"
)
json_fname = os.path.join(get_data_dir(), 'core_grid.json')
self.core_dataset = dc.data.NumpyDataset.from_json(json_fname)
def __init__(self,
sixty_four_bits: bool = True,
pocket_finder: Optional[BindingPocketFinder] = None):
"""Initializes Vina Pose Generator
Parameters
----------
sixty_four_bits: bool, optional (default True)
Specifies whether this is a 64-bit machine. Needed to download
the correct executable.
pocket_finder: BindingPocketFinder, optional (default None)
If specified should be an instance of
`dc.dock.BindingPocketFinder`.
"""
data_dir = get_data_dir()
if platform.system() == 'Linux':
url = "http://vina.scripps.edu/download/autodock_vina_1_1_2_linux_x86.tgz"
filename = "autodock_vina_1_1_2_linux_x86.tgz"
dirname = "autodock_vina_1_1_2_linux_x86"
self.vina_dir = os.path.join(data_dir, dirname)
self.vina_cmd = os.path.join(self.vina_dir, "bin/vina")
elif platform.system() == 'Darwin':
if sixty_four_bits:
url = "http://vina.scripps.edu/download/autodock_vina_1_1_2_mac_64bit.tar.gz"
filename = "autodock_vina_1_1_2_mac_64bit.tar.gz"
dirname = "autodock_vina_1_1_2_mac_catalina_64bit"
else:
url = "http://vina.scripps.edu/download/autodock_vina_1_1_2_mac.tgz"
filename = "autodock_vina_1_1_2_mac.tgz"
dirname = "autodock_vina_1_1_2_mac"
self.vina_dir = os.path.join(data_dir, dirname)
self.vina_cmd = os.path.join(self.vina_dir, "bin/vina")
elif platform.system() == 'Windows':
url = "http://vina.scripps.edu/download/autodock_vina_1_1_2_win32.msi"
filename = "autodock_vina_1_1_2_win32.msi"
self.vina_dir = "\\Program Files (x86)\\The Scripps Research Institute\\Vina"
self.vina_cmd = os.path.join(self.vina_dir, "vina.exe")
else:
raise ValueError(
"Unknown operating system. Try using a cloud platform to run this code instead."
)
self.pocket_finder = pocket_finder
if not os.path.exists(self.vina_dir):
logger.info("Vina not available. Downloading")
download_url(url, data_dir)
downloaded_file = os.path.join(data_dir, filename)
logger.info("Downloaded Vina. Extracting")
if platform.system() == 'Windows':
msi_cmd = "msiexec /i %s" % downloaded_file
check_output(msi_cmd.split())
else:
with tarfile.open(downloaded_file) as tar:
tar.extractall(data_dir)
logger.info("Cleanup: removing downloaded vina tar.gz")
os.remove(downloaded_file)
def __init__(self,
pretrain_model_path: Optional[str] = None,
radius: int = 1,
unseen: str = 'UNK',
gather_method: str = 'sum'):
"""
Paremeters
----------
pretrain_file: str, optional
The path for pretrained model. If this value is None, we use the model which is put on
github repository (https://github.com/samoturk/mol2vec/tree/master/examples/models).
The model is trained on 20 million compounds downloaded from ZINC.
radius: int, optional (default 1)
The fingerprint radius. The default value was used to train the model which is put on
github repository.
unseen: str, optional (default 'UNK')
The string to used to replace uncommon words/identifiers while training.
gather_method: str, optional (default 'sum')
How to aggregate vectors of identifiers are extracted from Mol2vec.
'sum' or 'mean' is supported.
"""
try:
from gensim.models import word2vec
from mol2vec.features import mol2alt_sentence, sentences2vec
except ModuleNotFoundError:
raise ValueError("This class requires mol2vec to be installed.")
self.radius = radius
self.unseen = unseen
self.gather_method = gather_method
self.sentences2vec = sentences2vec
self.mol2alt_sentence = mol2alt_sentence
if pretrain_model_path is None:
data_dir = get_data_dir()
pretrain_model_path = path.join(data_dir,
'mol2vec_model_300dim.pkl')
if not path.exists(pretrain_model_path):
targz_file = path.join(data_dir, 'mol2vec_model_300dim.tar.gz')
if not path.exists(targz_file):
download_url(DEFAULT_PRETRAINED_MODEL_URL, data_dir)
untargz_file(
path.join(data_dir, 'mol2vec_model_300dim.tar.gz'),
data_dir)
# load pretrained models
self.model = word2vec.Word2Vec.load(pretrain_model_path)
def create_gene_ontology(feature_mapping,
outputs_per_feature=0.3,
min_outputs=20,
min_node_features=6,
omit_redundant_nodes=True,
ontology_file=None):
"""Create a tree of OntologyNodes describing the Gene Ontology classification.
See http://geneontology.org/ for details about the Gene Ontology classification.
Parameters
----------
feature_mapping: dict
defines the mapping of features to GO categories. Each key should be a
feature ID. The corresponding value should be a list of strings, giving the
unique identifiers of all GO categories that feature belongs to.
outputs_per_feature: float
the number of outputs for each node is set to this value times the total
number of features the node contains (including all subnodes)
min_outputs: int
the minimum number of outputs for any node
min_node_features: int
the minimum number of features corresponding to a node (including all its
subnodes). If a category has fewer features than this, no node is create
for it. Instead, its features are added directly to its parent node.
omit_redundant_nodes: bool
if True, a node will be omitted if it has only one child node and does not
directly directly correspond to any features
ontology_file: str
the path to a Gene Ontology OBO file defining the ontology. If this is
omitted, the most recent version of the ontology is downloaded from the GO
website.
"""
# If necessary, download the file defining the ontology.
if ontology_file is None:
ontology_file = os.path.join(get_data_dir(), 'go-basic.obo')
if not os.path.isfile(ontology_file):
download_url('http://purl.obolibrary.org/obo/go/go-basic.obo')
# Parse the ontology definition and create a list of terms.
terms = []
term = None
with open(ontology_file) as input:
for line in input:
if line.startswith('[Term]'):
if term is not None:
terms.append(term)
term = {'parents': []}
elif line.startswith('[Typedef]'):
if term is not None:
terms.append(term)
term = None
elif line.startswith('id:') and term is not None:
term['id'] = line.split()[1]
elif line.startswith('name:') and term is not None:
term['name'] = line[5:].strip()
elif line.startswith('is_a:') and term is not None:
term['parents'].append(line.split()[1])
elif line.startswith('is_obsolete:'):
if line.split()[1] == 'true':
term = None
if term is not None:
terms.append(term)
# Create OntologyNode objects for all the terms.
nodes = {}
for term in terms:
nodes[term['id']] = OntologyNode(term['id'], 0, name=term['name'])
# Assign parent-child relationships between nodes, and identify root nodes.
roots = []
for term in terms:
node = nodes[term['id']]
for parent in term['parents']:
nodes[parent].children.append(node)
if len(term['parents']) == 0:
roots.append(node)
# Create a single root node that combines the three GO roots.
root = OntologyNode('GO', 0, name='Gene Ontology Root Node', children=roots)
# Assign features to nodes.
for feature_id in feature_mapping:
for node_id in feature_mapping[feature_id]:
nodes[node_id].feature_ids.append(feature_id)
# Count the number of features within each node. Eliminate nodes with too few
# features and set the number of outputs for each one.
def count_features(node):
self_features = set(node.feature_ids)
all_features = set(node.feature_ids)
for i, child in enumerate(node.children[:]):
child_features = count_features(child)
all_features.update(child_features)
if len(child_features) < min_node_features:
node.children.remove(child)
self_features.update(child.feature_ids)
if omit_redundant_nodes and len(
node.children) == 1 and len(self_features) == 0:
self_features = node.children[0].feature_ids
node.children = node.children[0].children
n_features = len(self_features)
if n_features > len(node.feature_ids):
node.feature_ids = list(self_features)
node.n_outputs = max(min_outputs,
math.ceil(outputs_per_feature * n_features))
return all_features
count_features(root)
return root