def from_json(cls, json: JSON) -> 'ProjectPublication':
title = lookup(json, 'title', 'publication_title')
url = lookup(json, 'url', 'publication_url', default=None)
return cls(title=title,
url=url,
official_hca=json.get('official_hca_publication'),
doi=json.get('doi'))
def __init__(self, json: JSON):
super().__init__(json)
content = json.get('content', json)
self.genus_species = {ontology_label(gs) for gs in content['genus_species']}
self.diseases = {ontology_label(d) for d in lookup(content, 'diseases', 'disease', default=[]) if d}
self.organism_age = content.get('organism_age')
self.organism_age_unit = ontology_label(content.get('organism_age_unit'), default=None)
self.sex = lookup(content, 'sex', 'biological_sex')
def __init__(self, json: JSON):
super().__init__(json)
content = json.get('content', json)
preservation_storage = content.get('preservation_storage')
self.storage_method = preservation_storage.get('storage_method') if preservation_storage else None
self.preservation_method = preservation_storage.get('preservation_method') if preservation_storage else None
self.diseases = {ontology_label(d) for d in lookup(content, 'diseases', 'disease', default=[]) if d}
self.organ = ontology_label(content.get('organ'), default=None)
organ_parts = lookup(content, 'organ_parts', 'organ_part', default=[])
if not isinstance(organ_parts, list):
organ_parts = [organ_parts]
self.organ_parts = {ontology_label(d) for d in organ_parts if d}
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
self.estimated_cell_count = lookup(content,
'estimated_cell_count',
'total_estimated_cells',
default=None)
self.selected_cell_types = {
ontology_label(sct)
for sct in lookup(content,
'selected_cell_types',
'selected_cell_type',
default=[])
}
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
self.type = lookup(content, 'type', 'cell_line_type')
self.model_organ = ontology_label(content.get('model_organ'),
default=None)
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
core = content['project_core']
self.project_short_name = lookup(core, 'project_short_name',
'project_shortname')
self.project_title = core['project_title']
self.project_description = core.get('project_description')
self.publications = set(
ProjectPublication.from_json(publication)
for publication in content.get('publications', []))
self.contributors = {
ProjectContact.from_json(contributor)
for contributor in content.get('contributors', [])
}
self.supplementary_links = set(content.get('supplementary_links', []))
self.estimated_cell_count = content.get('estimated_cell_count')
accessions = set()
for name, value in content.items():
prefix, _, suffix = name.rpartition('_')
if suffix == 'accessions':
assert prefix, name
assert isinstance(value, list)
accessions.update(
Accession(namespace=prefix, accession=v) for v in value)
self.accessions = accessions
def from_json(cls, json: JSON) -> Iterable['Link']:
if 'source_id' in json:
# v5
yield cls(source_id=UUID4(json['source_id']),
source_type=json['source_type'],
destination_id=UUID4(json['destination_id']),
destination_type=json['destination_type'])
else:
# vx
process_id = UUID4(json['process'])
for source_id in json['inputs']:
yield cls(source_id=UUID4(source_id),
source_type=json['input_type'],
destination_id=process_id,
destination_type='process')
for destination_id in json['outputs']:
yield cls(source_id=process_id,
source_type='process',
destination_id=UUID4(destination_id),
destination_type=json['output_type'])
for protocol in json['protocols']:
yield cls(source_id=process_id,
source_type='process',
destination_id=UUID4(protocol['protocol_id']),
destination_type=lookup(protocol, 'type', 'protocol_type'))
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
self.probe = [
ImagingProbe.from_json(probe)
for probe in lookup(content, 'probe', 'target', default=[])
]
def __init__(self, json: JSON):
super().__init__(json)
content = json.get('content', json)
self.genus_species = {gs['text'] for gs in content['genus_species']}
self.disease = {d['text'] for d in content.get('disease', []) if d}
self.organism_age = content.get('organism_age')
self.organism_age_unit = content.get('organism_age_unit',
{}).get('text')
self.sex = lookup(content, 'sex', 'biological_sex')
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
temp = lookup(content, 'library_construction_method',
'library_construction_approach')
self.library_construction_method = ontology_label(temp) if isinstance(
temp, dict) else temp
self.nucleic_acid_source = content.get('nucleic_acid_source')
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
self.model_organ = ontology_label(lookup(content, 'model_organ',
'model_for_organ'),
default=None)
self.model_organ_part = ontology_label(content.get('model_organ_part'),
default=None)
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
self.genus_species = {
ontology_label(gs)
for gs in content['genus_species']
}
self.diseases = {
ontology_label(d)
for d in lookup(content, 'diseases', 'disease', default=[]) if d
}
self.organism_age = content.get('organism_age')
self.organism_age_unit = ontology_label(
content.get('organism_age_unit'), default=None)
self.sex = lookup(content, 'sex', 'biological_sex')
self.development_stage = ontology_label(
content.get('development_stage'), default=None)
def __init__(self, json: JSON, manifest: Mapping[str, ManifestEntry]):
super().__init__(json)
content = json.get('content', json)
core = content['file_core']
self.format = lookup(core, 'format', 'file_format')
self.manifest_entry = manifest[core['file_name']]
self.content_description = {ontology_label(cd) for cd in core.get('content_description', [])}
self.from_processes = {}
self.to_processes = {}
def from_json(cls, json: JSON) -> 'ProjectContact':
project_role = json.get('project_role')
project_role = ontology_label(project_role) if isinstance(project_role, dict) else project_role
return cls(name=lookup(json, 'name', 'contact_name'),
email=json.get('email'),
institution=json.get('institution'),
laboratory=json.get('laboratory'),
corresponding_contributor=json.get('corresponding_contributor'),
project_role=project_role)
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
self.instrument_manufacturer_model = ontology_label(
content.get('instrument_manufacturer_model'), default=None)
self.paired_end = lookup(content,
'paired_end',
'paired_ends',
default=None)
def __init__(self, json: JSON) -> None:
super().__init__(json)
content = json.get('content', json)
core = content['project_core']
self.project_short_name = lookup(core, 'project_short_name',
'project_shortname')
self.laboratory_names = {
c.get('laboratory')
for c in content['contributors']
} - {None}
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry]) -> None:
super().__init__()
self.json = json
self.metadata_manifest_entry = metadata_manifest_entry
provenance = json.get('hca_ingest') or json['provenance']
self.document_id = UUID4(provenance['document_id'])
# Some older DCP/1 bundles use different UUIDs in the manifest and
# metadata.
# noinspection PyUnreachableCode
if False and self.metadata_manifest_entry is not None:
assert self.document_id == self.metadata_manifest_entry.uuid
self.submitter_id = provenance.get('submitter_id')
submission_date = lookup(provenance, 'submission_date',
'submissionDate')
self.submission_date = self._datetime(submission_date)
update_date = lookup(provenance,
'update_date',
'updateDate',
default=None)
self.update_date = self._optional_datetime(update_date)
def __init__(self, json: JSON) -> None:
super().__init__(json)
content = json.get('content', json)
core = content['project_core']
self.project_short_name = lookup(core, 'project_short_name', 'project_shortname')
self.project_title = core['project_title']
self.project_description = core.get('project_description')
self.publications = set(ProjectPublication.from_json(publication)
for publication in content.get('publications', []))
self.contributors = {ProjectContact.from_json(contributor) for contributor in content.get('contributors', [])}
self.insdc_project_accessions = set(content.get('insdc_project_accessions', []))
self.geo_series_accessions = set(content.get('geo_series_accessions', []))
self.array_express_accessions = set(content.get('array_express_accessions', []))
self.insdc_study_accessions = set(content.get('insdc_study_accessions', []))
self.supplementary_links = set(content.get('supplementary_links', []))
def __init__(self, json: JSON,
metadata_manifest_entry: Optional[ManifestEntry],
manifest: Mapping[str, ManifestEntry]):
super().__init__(json, metadata_manifest_entry)
content = json.get('content', json)
# '/' was once forbidden in file paths and was encoded with '!'. Now
# '/' is allowed and we force it in the metadata so that backwards
# compatibility is simplified downstream.
core = content['file_core']
core['file_name'] = core['file_name'].replace('!', '/')
self.format = lookup(core, 'format', 'file_format')
self.manifest_entry = manifest[core['file_name']]
self.content_description = {
ontology_label(cd)
for cd in core.get('content_description', [])
}
self.file_source = core.get('file_source')
self.from_processes = {}
self.to_processes = {}
def ontology_label(
ontology: Optional[Mapping[str, str]],
default: Union[str, None, LookupDefault] = LookupDefault.RAISE) -> str:
"""
Return the best-suited value from the given ontology dictionary.
>>> ontology_label({'ontology_label': '1', 'text': '2', 'ontology': '3'})
'1'
>>> ontology_label({'text': '2', 'ontology': '3'})
'2'
>>> ontology_label({'ontology': '3'})
'3'
>>> ontology_label({}, default=None)
>>> ontology_label({}, default='default')
'default'
>>> ontology_label(None, default=None)
>>> ontology_label(None, default='default')
'default'
>>> ontology_label({})
Traceback (most recent call last):
...
KeyError: 'ontology_label'
>>> ontology_label(None)
Traceback (most recent call last):
...
TypeError: 'NoneType' object is not subscriptable
"""
if ontology is None and default is not LookupDefault.RAISE:
return default
else:
return lookup(ontology,
'ontology_label',
'text',
'ontology',
default=default)
def __init__(self, json: JSON):
super().__init__(json)
content = json.get('content', json)
self.instrument_manufacturer_model = ontology_label(content.get('instrument_manufacturer_model'), default=None)
self.paired_end = lookup(content, 'paired_end', 'paired_ends', default=None)
def __init__(self, json: JSON) -> None:
super().__init__(json)
content = json.get('content', json)
temp = lookup(content, 'library_construction_method', 'library_construction_approach')
self.library_construction_method = ontology_label(temp) if isinstance(temp, dict) else temp
def __init__(self, json: JSON) -> None:
super().__init__(json)
content = json.get('content', json)
self.model_organ = ontology_label(lookup(content, 'model_organ', 'model_for_organ'), default=None)
self.model_organ_part = ontology_label(content.get('model_organ_part'), default=None)
def __init__(self, json: JSON) -> None:
super().__init__(json)
content = json.get('content', json)
self.type = lookup(content, 'type', 'cell_line_type')
self.model_organ = ontology_label(content.get('model_organ'), default=None)
def __init__(self, json: JSON) -> None:
super().__init__(json)
content = json.get('content', json)
self.estimated_cell_count = lookup(content, 'estimated_cell_count', 'total_estimated_cells', default=None)
self.selected_cell_types = {ontology_label(sct) for sct in
lookup(content, 'selected_cell_types', 'selected_cell_type', default=[])}
def from_json(cls, json: JSON,
schema_version: Tuple[int]) -> Iterable['Link']:
if 'source_id' in json:
# DCP/1 v5 (obsolete)
yield cls(source_id=UUID4(json['source_id']),
source_type=json['source_type'],
destination_id=UUID4(json['destination_id']),
destination_type=json['destination_type'])
elif schema_version[0] == 1:
# DCP/1 vx (current)
process_id = UUID4(json['process'])
for source_id in json['inputs']:
yield cls(source_id=UUID4(source_id),
source_type=json['input_type'],
destination_id=process_id,
destination_type='process')
for destination_id in json['outputs']:
yield cls(source_id=process_id,
source_type='process',
destination_id=UUID4(destination_id),
destination_type=json['output_type'])
for protocol in json['protocols']:
yield cls(source_id=process_id,
source_type='process',
destination_id=UUID4(protocol['protocol_id']),
destination_type=lookup(protocol, 'type',
'protocol_type'))
elif schema_version[0] in (2, 3):
# DCP/2 (current)
link_type = json['link_type']
if link_type == 'process_link':
process_id = UUID4(json['process_id'])
process_type = json['process_type']
for input_ in json['inputs']:
yield cls(link_type=link_type,
source_id=UUID4(input_['input_id']),
source_type=input_['input_type'],
destination_id=process_id,
destination_type=process_type)
for output in json['outputs']:
yield cls(link_type=link_type,
source_id=process_id,
source_type=process_type,
destination_id=UUID4(output['output_id']),
destination_type=output['output_type'])
for protocol in json['protocols']:
yield cls(link_type=link_type,
source_id=process_id,
source_type=process_type,
destination_id=UUID4(protocol['protocol_id']),
destination_type=protocol['protocol_type'])
elif link_type == 'supplementary_file_link':
entity = json['entity']
for supp_file in json['files']:
yield cls(link_type=link_type,
source_id=UUID4(entity['entity_id']),
source_type=entity['entity_type'],
destination_id=UUID4(supp_file['file_id']),
destination_type=supp_file['file_type'])
else:
assert False, f'Unknown link_type {link_type}'
else:
assert False, f'Unknown schema_version {schema_version}'
def from_json(cls, json: JSON) -> 'ProjectPublication':
title = lookup(json, 'title', 'publication_title')
url = lookup(json, 'url', 'publication_url', default=None)
return cls(title=title, url=url)
