def related(term, level=1, create_session=None):
'''
for relname, to_term in related(term, create_session={backend.create_session}):
....
Parameters
----------
term : str or unicode
GO term
level : int, optional
Maximum recursion level. Use ``-1`` for infinite recursion
create_session : callable
Returns
-------
iterator
'''
if create_session is None:
from waldo.backend import create_session
session = create_session()
seen = set()
def recurse(term, level):
if level == 0:
return
q = session\
.query(TermRelationship)\
.filter_by(from_term=term)
for rel in q.all():
if rel.to_term not in seen:
seen.add(rel.to_term)
recurse(rel.to_term, level-1)
recurse(term, level)
for r in seen:
yield r
def test_oop():
# This just checks that all methods exist &c
ensemblgene = 'ENSMUSG00000064345'
session = backend.create_session()
for module in (waldo.uniprot, waldo.mgi, waldo.locate):
name = module.retrieve.from_ensembl_gene_id(ensemblgene, session)
entry = module.retrieve.retrieve_entry(name, session)
if name is not None and entry is not None:
if module.name == 'LOCATE':
evidence = '<br />'.join(
['<a href="http://locate.imb.uq.edu.au/data_images/%s/%s"><img height="50" width="50" src="http://locate.imb.uq.edu.au/data_images/%s/%s" /></a>' %
(img.filename[0:3], img.filename, img.filename[0:3], img.filename) for img in entry.images]),
else:
try:
evidence = '<br />'.join((annot.evidence or 'None') for annot in entry.go_annotations),
except:
evidence = '-'
d = {
'protein': entry.name,
'organism' : '; '.join(entry.organisms),
'celltype': '-',
'condition':'-',
'location': '; '.join(module.retrieve.retrieve_go_annotations(entry.internal_id)),
'references': '<br />'.join([paper.title for paper in entry.references]),
'evidence' : evidence,
'evidence_code' : '<br />'.join([go_annot.evidence_code for go_annot in entry.go_annotations]),
'source':'<a href="%s">%s</a>' % (module.retrieve.gen_url(entry.name),module.name),
}
def vocabulary(go_id, session=None):
'''
voc = vocabulary(go_id, session={new session})
Return the vocabulary to which the `go_id` belongs.
Parameters
----------
go_id : str
A GO id (e.g., GO:123456789)
session : SQLAlchemy session object, optional
Returns
-------
voc: str
Vocabulary (one of 'cellular_component', 'molecular_process', 'biological_function')
None if not found.
'''
if session is None:
session = create_session()
entry = session.query(models.Term) \
.filter(models.Term.id == go_id) \
.first()
if entry:
return entry.namespace
def term_to_id(go_term, session=None):
'''
go_id = term_to_id(go_term, session={waldo.backend.create_session()})
For a GO term, returns the corresponding ID.
If no ID is found, returns the original term.
Parameters
----------
term : string
A GO term (e.g. "mitochondrion")
session : SQLAlchemy session, optional
Returns
-------
go_id : string
A GO ID (e.g., GO:0005739)
'''
if session is None: session = create_session()
term = session.query(Term).filter(Term.name == go_term).first()
if term is None:
# if there's no mapping, return the original id
return go_term
return term.id
def map_to_goslim(go_id, set_name, session=None):
'''
go_slim_term = map_to_goslim(go_id, set_name, session={backend.create_session()})
Map a GO term to a GO slim term in sub-vocabulary `set_name`
Parameters
----------
go_id : string
GO id
set_name : string
name of GO Slim to use
session : sqlalchemy session, optional
sqlalchemy session (default : `backend.create_session()`)
Returns
-------
go_slim_term : str or None
GO Slim Term corresponding to input `go_id`
'''
if set_name not in _known_slimsets:
raise KeyError('waldo.goslim.map_to_goslim: Unknown GO Slim `%s` (known are %s)' % (set_name, _known_slimsets))
if session is None:
from waldo.backend import create_session
session = create_session()
entry = (session.query(SlimTerm)
.join(SlimMapping)
.join(SlimSet)
.filter(SlimSet.id == set_name)
.filter(SlimMapping.full_id == go_id)
.first())
if entry is None: return None
return entry.name
def retrieve_orthologs(ensembl_id, session=None):
'''
orthologs = retrieve_orthologs(ensembl_id, session={backend.create_session()})
Retrieve orthologs by Ensembl peptide id.
Parameters
----------
ensembl_id : str
Ensembl Peptide ID
session : SQLAlchemy session, optional
to use (default: call backend.create_session())
Returns
-------
orthologs : list of str or None
List of Ensembl IDs. If the protein is not found, None is returned. If
the protein is found, but no orthologs are defined, an empty list is
returned.
'''
if session is None: session = backend.create_session()
entry = session.query(NogEntry).filter(NogEntry.ensembl_peptide == ensembl_id).first()
if entry is None:
return None
entries = session.query(NogEntry).filter(NogEntry.orthologous_group == entry.orthologous_group).all()
entries = [e.ensembl_peptide for e in entries]
entries = [e for e in entries if e != ensembl_id]
return entries
def retrieve_go_annotations(id, session=None):
'''
go_ids = retrieve_go_annotations(name, session={backend.create_session()})
Retrieve GO ids by LOCATE identifier.
Parameters
----------
id : str
LOCATE protein uid
session : SQLAlchemy session object, optional
session to use (default: call backend.create_session())
Returns
-------
go_ids : list of str
go terms (of the form "GO:00...")
'''
if session is None: session = backend.create_session()
entry = session.query(Entry).filter(Entry.id == id).first()
locations = set([ann.go_id for ann in entry.go_annotations])
for predict in entry.predictions:
locations.update(predict.go_id.split(';'))
for reference in entry.references:
for location in reference.locations:
locations.update(location.go_id.split(';'))
for annotation in entry.go_annotations:
for location in annotation.locations:
locations.update(location.go_id.split(';'))
return list(locations)
def retrieve_entry(id, session=None):
'''
entry = retrieve_entry(id, session={backend.create_session()})
Retrieve an Entry object based on its ID
Parameters
----------
id : LOCATE database ID
session : SQLAlchemy session to use (default: create a new one)
Returns
-------
entry : A models.Entry object
'''
if session is None: session = backend.create_session()
return session.query(Entry).filter(Entry.id == id).first()
def id_to_term(go_id, session=None):
'''
term = id_to_term(go_id, session={waldo.backend.create_session()})
For a GO ID, returns the corresponding term.
Parameters
----------
go_id : str
A GO ID (e.g., GO:0005739)
session : SQLAlchemy session object, optional
Returns
-------
term : str
A GO term (e.g. "mitochondrion")
'''
if session is None: session = create_session()
term = session.query(Term).filter(Term.id == go_id).first()
if term is None:
# if there's no mapping, return the original id
return go_id
else:
return term.name