def add_phenotype(self, ind_obj, phenotype_id):
"""Add a phenotype term to the case."""
if phenotype_id.startswith('HP:') or len(phenotype_id) == 7:
logger.debug('querying on HPO term')
hpo_results = phizz.query_hpo([phenotype_id])
else:
logger.debug('querying on OMIM term')
hpo_results = phizz.query_disease([phenotype_id])
added_terms = [] if hpo_results else None
existing_ids = set(term.phenotype_id for term in ind_obj.phenotypes)
for result in hpo_results:
if result['hpo_term'] not in existing_ids:
term = PhenotypeTerm(phenotype_id=result['hpo_term'],
description=result['description'])
logger.info('adding new HPO term: %s', term.phenotype_id)
ind_obj.phenotypes.append(term)
added_terms.append(term)
logger.debug('storing new HPO terms')
self.save()
if added_terms is not None and len(added_terms) > 0:
for case_obj in ind_obj.cases:
self.update_hpolist(case_obj)
return added_terms
def add_phenotype(self, ind_obj, phenotype_id):
"""Add a phenotype term to the case."""
if phenotype_id.startswith('HP:') or len(phenotype_id) == 7:
logger.debug('querying on HPO term')
hpo_results = phizz.query_hpo([phenotype_id])
else:
logger.debug('querying on OMIM term')
hpo_results = phizz.query_disease([phenotype_id])
added_terms = [] if hpo_results else None
existing_ids = set(term.phenotype_id for term in ind_obj.phenotypes)
for result in hpo_results:
if result['hpo_term'] not in existing_ids:
term = PhenotypeTerm(phenotype_id=result['hpo_term'],
description=result['description'])
logger.info('adding new HPO term: %s', term.phenotype_id)
ind_obj.phenotypes.append(term)
added_terms.append(term)
logger.debug('storing new HPO terms')
self.save()
if added_terms is not None and len(added_terms) > 0:
for case_obj in ind_obj.cases:
self.update_hpolist(case_obj)
return added_terms
def get_hpo(variant_str):
samples=get_samples(variant_str)
#chrom,pos,ref,alt,=str(variant_str.strip()).split('-')
d=csv.DictReader(file('/data/uclex_data/UCLexInfo/uclex-samples.csv','r'),delimiter=',')
hpo=[]
for r in d:
if r['sample'] not in samples: continue
pheno=r['phenotype']
print((r['sample'],pheno,))
if pheno.startswith('HP'):
hpo+=[phizz.query_hpo([pheno])]
elif pheno.startswith('MIM'):
hpo+=[phizz.query_disease([pheno])]
return(hpo)
def get_hpo(variant_str):
samples=get_samples(variant_str)
#chrom,pos,ref,alt,=str(variant_str.strip()).split('-')
d=csv.DictReader(file('/data/uclex_data/UCLexInfo/uclex-samples.csv','r'),delimiter=',')
hpo=[]
for r in d:
if r['sample'] not in samples: continue
pheno=r['phenotype']
print((r['sample'],pheno,))
if pheno.startswith('HP'):
hpo+=[phizz.query_hpo([pheno])]
elif pheno.startswith('MIM'):
hpo+=[phizz.query_disease([pheno])]
return(hpo)
def phenotypes(variant):
for s in variant['HOM']+variant['HET']:
for hpo in phenotips.patient_hpo(eid=s,auth=args.login):
if hpo.startswith('HP'):
variant['HPO']=variant.get('HPO',[])+phizz.query_hpo([hpo])
elif hpo.startswith('MIM'):
variant['HPO']=variant.get('HPO',[])+phizz.query_disease([hpo])
#for hpo in variant['HPO']: variant['HPO'+=get_ancestors(hpo['hpo_term'])
if 'HPO' in variant:
#variant['HPO']=list(set([h['hpo_term'] for h in variant['HPO']]))
variant['HPO']=[h['hpo_term'] for h in variant['HPO']]
else:
variant['HPO']=[]
return(variant)
def phenotypes(variant):
for s in variant['HOM'] + variant['HET']:
for hpo in phenotips.patient_hpo(eid=s, auth=args.login):
if hpo.startswith('HP'):
variant['HPO'] = variant.get('HPO', []) + phizz.query_hpo(
[hpo])
elif hpo.startswith('MIM'):
variant['HPO'] = variant.get('HPO', []) + phizz.query_disease(
[hpo])
#for hpo in variant['HPO']: variant['HPO'+=get_ancestors(hpo['hpo_term'])
if 'HPO' in variant:
#variant['HPO']=list(set([h['hpo_term'] for h in variant['HPO']]))
variant['HPO'] = [h['hpo_term'] for h in variant['HPO']]
else:
variant['HPO'] = []
return (variant)
def add_phenotype(self, ind_obj, phenotype_id):
"""Add a phenotype term to the case."""
if phenotype_id.startswith("HP:") or len(phenotype_id) == 7:
logger.debug("querying on HPO term")
hpo_results = phizz.query_hpo([phenotype_id])
else:
logger.debug("querying on OMIM term")
hpo_results = phizz.query_disease([phenotype_id])
added_terms = []
existing_ids = set(term.phenotype_id for term in ind_obj.phenotypes)
for result in hpo_results:
if result["hpo_term"] not in existing_ids:
term = PhenotypeTerm(phenotype_id=result["hpo_term"], description=result["description"])
logger.info("adding new HPO term: %s", term.phenotype_id)
ind_obj.phenotypes.append(term)
added_terms.append(term)
logger.debug("storing new HPO terms")
self.save()
return added_terms
def add_phenotype(self, institute, case, user, link, hpo_term=None,
omim_term=None):
"""Add a new phenotype term to a case
Create a phenotype term and event with the given information
Args:
institute (Institute): A Institute object
case (Case): Case object
user (User): A User object
link (str): The url to be used in the event
hpo_term (str): A hpo id
omim_term (str): A omim id
"""
try:
if hpo_term:
logger.debug("Fetching info for hpo term {0}".format(hpo_term))
hpo_results = phizz.query_hpo([hpo_term])
elif omim_term:
logger.debug("Fetching info for mim term {0}".format(omim_term))
hpo_results = phizz.query_disease([omim_term])
else:
raise ValueError('Must supply either hpo or omim term')
logger.debug("Got result {0}".format(
', '.join(res['hpo_term'] for res in hpo_results)))
except ValueError as e:
#TODO Should ve raise a more proper exception here?
raise e
phenotype_terms = []
for hpo_result in hpo_results:
phenotype_name = hpo_result['hpo_term']
description = hpo_result['description']
phenotype_term = PhenotypeTerm(
phenotype_id=phenotype_name,
feature=description
)
phenotype_terms.append(phenotype_term)
logger.info("Append the phenotype term {0} to case {1}".format(
phenotype_name, case.display_name
))
case.phenotype_terms.append(phenotype_term)
logger.info("Creating event for adding phenotype term for case"\
" {0}".format(case.display_name))
self.create_event(
institute=institute,
case=case,
user=user,
link=link,
category='case',
verb='add_phenotype',
subject=case.display_name,
content = phenotype_name
)
case.save()
logger.debug("Case updated")
def get_awesomebar_result(db, query):
"""
Similar to the above, but this is after a user types enter
We need to figure out what they meant - could be gene, variant, region
Return tuple of (datatype, identifier)
Where datatype is one of 'gene', 'variant', or 'region'
And identifier is one of:
- ensembl ID for gene
- variant ID string for variant (eg. 1-1000-A-T)
- region ID string for region (eg. 1-1000-2000)
Follow these steps:
- if query is an ensembl ID, return it
- if a gene symbol, return that gene's ensembl ID
- if an RSID, return that variant's string
Finally, note that we don't return the whole object here - only it's identifier.
This could be important for performance later
"""
query = query.strip()
print 'Query: %s' % query
if query.startswith('HP:'):
description=phizz.query_hpo([query])
#description=hpo_db.hpo.find_one({'hpo_id':query})
return 'hpo', query
if query.startswith('MIM'):
disease=phizz.query_disease([query])
return 'mim', query
# Variant
variant = orm.get_variants_by_rsid(db, query.lower())
if variant:
if len(variant) == 1:
return 'variant', variant[0]['variant_id']
else:
return 'dbsnp_variant_set', variant[0]['rsid']
variant = get_variants_from_dbsnp(db, query.lower())
if variant:
return 'variant', variant[0]['variant_id']
# variant = get_variant(db, )
# TODO - https://github.com/brettpthomas/exac_browser/issues/14
gene = get_gene_by_name(db, query)
if gene:
return 'gene', gene['gene_id']
# From here out, all should be uppercase (gene, tx, region, variant_id)
query = query.upper()
gene = get_gene_by_name(db, query)
if gene:
return 'gene', gene['gene_id']
# Ensembl formatted queries
if query.startswith('ENS'):
# Gene
gene = get_gene(db, query)
if gene:
return 'gene', gene['gene_id']
# Transcript
transcript = get_transcript(db, query)
if transcript:
return 'transcript', transcript['transcript_id']
# From here on out, only region queries
if query.startswith('CHR'):
query = query.lstrip('CHR')
# Region
m = R1.match(query)
if m:
if int(m.group(3)) < int(m.group(2)):
return 'region', 'invalid'
return 'region', '{}-{}-{}'.format(m.group(1), m.group(2), m.group(3))
m = R2.match(query)
if m:
return 'region', '{}-{}-{}'.format(m.group(1), m.group(2), m.group(2))
m = R3.match(query)
if m:
return 'region', '{}'.format(m.group(1))
m = R4.match(query)
if m:
return 'variant', '{}-{}-{}-{}'.format(m.group(1), m.group(2), m.group(3), m.group(4))
return 'not_found', query
path = '/slms/UGI/vm_exports/vyp/phenotips/uclex_files/hp.obo'
ontology = Ontology(path)
hpo_graph = ontology.get_graph()
#print('\t'.join(['external_id','mode_of_inheritance','hpo_features','hpo_ancestors']))
for l in sys.stdin.readlines():
io = StringIO(l)
p=json.load(io)
external_id=p['external_id']
mode_of_inheritance= [i['id'] for i in p.get('global_mode_of_inheritance',[])]
hpo=[]
for f in p.get('features',[]):
if f['id'].startswith('MIM'):
hpo+=[x['hpo_term'] for x in phizz.query_disease([f['id']])]
else:
hpo+=[f['id']]
if hpo:
hpo_ancestors = list(set(sum([list(hpo_graph.get_ancestors(h)) for h in hpo],[])))
else:
hpo_ancestors = []
#print(external_id,','.join(mode_of_inheritance),','.join(hpo),','.join(hpo_ancestors),sep='\t')
for h in list(set(hpo+hpo_ancestors)):
q=db.hpo.find_one({'hp_id':h})
if q is not None:
external_ids=q['external_ids']
external_ids=list(set([external_id]+external_ids))
r=db.hpo.update({'hp_id':h}, {'$set':{'external_ids':external_ids}},upsert=True)
print('update',h,external_ids)
else:
