def _process_concept(*, concept, node=None) -> bool:
"""Process a node JSON object."""
namespace = concept[NAMESPACE]
if namespace.lower() in {'text', 'fixme'}:
return False
prefix = normalize_prefix(namespace)
if prefix is None:
logger.warning('could not normalize namespace %s in concept %s', namespace, concept)
return False
concept[NAMESPACE] = prefix
identifier = concept.get(IDENTIFIER)
name = concept.get(NAME)
if identifier: # don't trust whatever was put for the name, even if it's available
map_success = _handle_identifier_not_name(concept=concept, prefix=prefix, identifier=identifier)
else:
map_success = _handle_name_and_not_identifier(concept=concept, prefix=prefix, name=name, node=node)
if not map_success:
return False
_remap_by_identifier(concept)
return True
def summarize_one(prefix: str):
"""Summarize the mappings."""
prefix = normalize_prefix(prefix)
in_df = summary_df.loc[summary_df[TARGET_PREFIX] == prefix,
[SOURCE_PREFIX, 'count']]
out_df = summary_df.loc[summary_df[SOURCE_PREFIX] == prefix,
[TARGET_PREFIX, 'count']]
return f'''
def _process_annotations(data, add_free_annotations: bool = False):
x = []
y = []
for prefix, names in data[ANNOTATIONS].items():
if prefix == 'CellLine':
efo_name_to_id = get_name_id_mapping('efo')
# clo_name_to_id = get_name_id_mapping('clo') # FIXME implement CLO import
for name, polarity in names.items():
prefix, identifier = 'efo', efo_name_to_id.get(name)
# if identifier is None:
# prefix, identifier = 'clo', clo_name_to_id.get(name)
if identifier is not None:
x.append((Entity(namespace=prefix, identifier=identifier, name=name), polarity))
else:
y.append((prefix, identifier, polarity))
elif prefix in _BEL_ANNOTATION_PREFIX_MAP:
prefix = _BEL_ANNOTATION_PREFIX_MAP[prefix]
_process_findable_annotations(x, y, prefix, names)
elif normalize_prefix(prefix):
prefix_norm = normalize_prefix(prefix)
_process_findable_annotations(x, y, prefix_norm, names)
else:
if prefix not in _UNHANDLED_ANNOTATION:
logger.warning('unhandled annotation: %s', prefix)
_UNHANDLED_ANNOTATION.add(prefix)
if isinstance(names, dict):
for name, polarity in names.items():
y.append((prefix, name, polarity))
else:
y.append((prefix, names, True))
data[ANNOTATIONS] = defaultdict(dict)
for entity, polarity in x:
data[ANNOTATIONS][entity.namespace][entity.identifier] = polarity
if add_free_annotations:
data['free_annotations'] = defaultdict(dict)
for prefix, name, polarity in y:
data['free_annotations'][prefix][name] = polarity
def get_remapping() -> Mapping[Tuple[str, str], Tuple[str, str, str]]:
"""Get a mapping from database/identifier pairs to famplex identifiers."""
df = _get_df()
rv = {}
for target_ns, target_id, source_id in df.values:
remapped_prefix = normalize_prefix(target_ns)
if remapped_prefix is None:
logger.debug('could not remap %s', target_ns)
else:
rv[remapped_prefix, target_id] = 'fplx', source_id, source_id
return rv
def _process_concept(
*,
concept,
node=None,
skip_namespaces: Optional[Collection[str]] = None) -> bool:
"""Process a node JSON object."""
namespace = concept[NAMESPACE]
if namespace.lower() in {'text', 'fixme'}:
return False
if skip_namespaces and namespace in skip_namespaces:
return True
prefix = normalize_prefix(namespace)
if prefix is None:
logger.warning('could not normalize namespace "%s" in concept "%s"',
namespace, concept)
return False
concept[NAMESPACE] = prefix
identifier = concept.get(IDENTIFIER)
name = concept.get(NAME)
if identifier: # don't trust whatever was put for the name, even if it's available
map_success = _handle_identifier_not_name(
concept=concept,
prefix=prefix,
identifier=identifier,
skip_namespaces=skip_namespaces)
if not map_success: # just in case the name gets put in the identifier
map_success = _handle_name_and_not_identifier(
concept=concept,
prefix=prefix,
name=identifier,
node=node,
skip_namespaces=skip_namespaces,
)
else:
map_success = _handle_name_and_not_identifier(
concept=concept,
prefix=prefix,
name=name,
node=node,
skip_namespaces=skip_namespaces,
)
if not map_success:
return False
_remap_by_identifier(concept)
return True
def _process_annotations(data, remove_ungrounded: bool = False) -> None:
"""Process the annotations in a PyBEL edge data dictionary."""
grounded_category_curie_polarity = []
ungrounded_category_name_polarity = []
for category, names in data[ANNOTATIONS].items():
if category == 'CellLine':
_namespaces = [
'efo',
# 'clo', # FIXME implement CLO import and add here
]
for name, polarity in names.items():
g_prefix, g_identifier, g_name = pyobo.ground(_namespaces, name)
if g_prefix and g_identifier:
grounded_category_curie_polarity.append((
category, Entity(namespace=g_prefix, identifier=g_identifier, name=g_name), polarity,
))
else:
ungrounded_category_name_polarity.append((category, name, polarity))
elif category in _BEL_ANNOTATION_PREFIX_MAP:
norm_prefix = _BEL_ANNOTATION_PREFIX_MAP[category]
norm_category = _BEL_ANNOTATION_PREFIX_CATEGORY_MAP.get(category, category)
for name, polarity in names.items():
_, identifier, _ = pyobo.ground(norm_prefix, name)
if identifier:
grounded_category_curie_polarity.append((
norm_category, Entity(namespace=norm_prefix, identifier=identifier, name=name), polarity,
))
else:
ungrounded_category_name_polarity.append((norm_category, name, polarity))
elif normalize_prefix(category):
norm_prefix = normalize_prefix(category)
for name, polarity in names.items():
_, identifier, _ = pyobo.ground(norm_prefix, name)
if identifier:
grounded_category_curie_polarity.append((
category, Entity(namespace=norm_prefix, identifier=identifier, name=name), polarity,
))
else:
ungrounded_category_name_polarity.append((category, name, polarity))
else:
if category not in _UNHANDLED_ANNOTATION:
logger.warning('unhandled annotation: %s', category)
_UNHANDLED_ANNOTATION.add(category)
if isinstance(names, dict):
for name, polarity in names.items():
ungrounded_category_name_polarity.append((category, name, polarity))
else:
ungrounded_category_name_polarity.append((category, names, True))
data[ANNOTATIONS] = defaultdict(dict)
for category, entity, polarity in grounded_category_curie_polarity:
data[ANNOTATIONS][category][entity.curie] = polarity
data[ANNOTATIONS] = dict(data[ANNOTATIONS])
if not remove_ungrounded and ungrounded_category_name_polarity:
data[FREE_ANNOTATIONS] = defaultdict(dict)
for category, name, polarity in ungrounded_category_name_polarity:
data[FREE_ANNOTATIONS][category][name] = polarity
data[FREE_ANNOTATIONS] = dict(data[FREE_ANNOTATIONS])
def summarize_one(prefix: str):
"""Summarize the mappings."""
prefix = normalize_prefix(prefix)
in_df = summary_df.loc[summary_df['target_ns'] == prefix, ['source_ns', 'count']]
out_df = summary_df.loc[summary_df['source_ns'] == prefix, ['target_ns', 'count']]
return f'''