def test_update(self):
"""
Test update investigator
"""
# Create studies and investigator
studies, investigator, kwargs = self.create_investigator()
# Add new study to investigator
s_new = Study(external_id='phs002', investigator_id=investigator.kf_id)
db.session.add(s_new)
db.session.commit()
# Check database
self.assertEqual(
3, len(Investigator.query.get(investigator.kf_id).studies))
self.assertIn(s_new,
Investigator.query.get(investigator.kf_id).studies)
# Add study to new investigator
i = Investigator(name='Alma')
i.studies = [s_new]
db.session.commit()
# Check database
self.assertEqual(
i.studies[0].kf_id,
Study.query.filter_by(external_id='phs002').one().kf_id)
self.assertNotIn(i, Investigator.query.get(investigator.kf_id).studies)
def create_investigator(self):
# Create investigator
investigator = Investigator(name='Adam',
institution='CHOP',
external_id='Invest_1')
# Create studies
studies = []
for i in range(2):
kwargs = {
'attribution': ('https://dbgap.ncbi.nlm.nih.gov/'
'aa/wga.cgi?view_pdf&stacc=phs000178.v9.p8'),
'external_id':
'phs00{}'.format(i),
'name':
'study_{}'.format(i),
'version':
'v1'
}
studies.append(Study(**kwargs))
# Add study to investigator
investigator.studies.extend(studies)
db.session.add(investigator)
db.session.commit()
return studies, investigator, kwargs
def test_foreign_key_constraint(self):
"""
Test that a investigator can be created without existing
reference Study. This checks foreign key constraint
"""
# Create investigator
i = Investigator(name='Adam', institution='CHOP')
db.session.add(i)
db.session.commit()
# Check for database
self.assertEqual(1, Investigator.query.count())
def _create_studies_investigators(self, total=None):
"""
Create studies and investigators
investigator.csv contains list of investigator name, instistuion name
and studies
"""
pref_file = open('dataservice/util/data_gen/investigator.csv', 'r')
reader = csv.reader(pref_file)
investigator_chosen_list = []
for row in reader:
investigator_chosen_list.append(row)
invest_study = random.choice(investigator_chosen_list)
# Create Investigators
min_investigators = 1
max_investigators = len(invest_study)
if not total:
total = random.randint(min_investigators, max_investigators)
investigators = []
for i in range(total):
kwargs = {'name': invest_study[0], 'institution': invest_study[1]}
inv = Investigator(**kwargs)
investigators.append(inv)
db.session.add(inv)
db.session.commit()
# Create Studies
studies = []
for i in range(total):
kwargs = {
'attribution': ('https://dbgap.ncbi.nlm.nih.gov/'
'aa/wga.cgi?view_pdf&stacc=phs000178.v9.p8'),
'external_id':
'phs00{}'.format(i),
'name':
invest_study[2],
'version':
'v1',
'investigator_id':
investigators[i].kf_id
}
study_files = self._create_study_files(random.randint(0, total))
s = Study(**kwargs, study_files=study_files)
studies.append(s)
db.session.add(s)
db.session.commit()
return studies
def _make_study(self, external_id='TEST-0001', include_nullables=True):
"""
Convenience method to create a study with a given source name
"""
inv = Investigator(name='donald duck')
db.session.add(inv)
db.session.flush()
body = {
'external_id': external_id,
'version': '1.0',
'release_status': 'Pending'
}
if include_nullables:
body.update({'investigator_id': inv.kf_id})
response = self.client.post(url_for(STUDY_LIST_URL),
headers=self._api_headers(),
data=json.dumps(body))
return response
def participants(client):
# Add a bunch of studies for pagination
for i in range(101):
s = Study(external_id='Study_{}'.format(i))
db.session.add(s)
for i in range(101):
ca = CavaticaApp(name='app', revision=0)
db.session.add(ca)
# Add a bunch of study files
s0 = Study.query.filter_by(external_id='Study_0').one()
s1 = Study.query.filter_by(external_id='Study_1').one()
for i in range(101):
sf = StudyFile(file_name='blah', study_id=s0.kf_id)
db.session.add(sf)
# Add a bunch of investigators
for _ in range(102):
inv = Investigator(name='test')
inv.studies.extend([s0, s1])
db.session.add(inv)
# Add a bunch of families
families = []
for i in range(101):
families.append(Family(external_id='Family_{}'.format(i)))
db.session.add_all(families)
db.session.flush()
participants = []
f0 = Family.query.filter_by(external_id='Family_0').one()
f1 = Family.query.filter_by(external_id='Family_1').one()
seq_cen = None
for i in range(102):
f = f0 if i < 50 else f1
s = s0 if i < 50 else s1
data = {
'external_id': "test",
'is_proband': True,
'race': 'Asian',
'ethnicity': 'Hispanic or Latino',
'diagnosis_category': 'Cancer',
'gender': 'Male'
}
p = Participant(**data, study_id=s.kf_id, family_id=f.kf_id)
diag = Diagnosis()
p.diagnoses = [diag]
outcome = Outcome()
p.outcomes = [outcome]
phen = Phenotype()
p.phenotypes = [phen]
participants.append(p)
db.session.add(p)
db.session.flush()
seq_data = {
'external_id': 'Seq_0',
'experiment_strategy': 'WXS',
'library_name': 'Test_library_name_1',
'library_strand': 'Unstranded',
'is_paired_end': False,
'platform': 'Test_platform_name_1'
}
gf_kwargs = {
'external_id': 'gf_0',
'file_name': 'hg38.fq',
'data_type': 'Aligned Reads',
'file_format': 'fastq',
'size': 1000,
'urls': ['s3://bucket/key'],
'hashes': {
'md5': str(uuid.uuid4())
},
'controlled_access': False
}
seq_cen = SequencingCenter.query.filter_by(name="Baylor")\
.one_or_none()
if seq_cen is None:
seq_cen = SequencingCenter(external_id='SC_0', name="Baylor")
db.session.add(seq_cen)
db.session.flush()
seq_exp = SequencingExperiment(**seq_data,
sequencing_center_id=seq_cen.kf_id)
db.session.add(seq_exp)
samp = Biospecimen(analyte_type='an analyte',
sequencing_center_id=seq_cen.kf_id,
participant=p)
db.session.add(samp)
p.biospecimens = [samp]
gf = GenomicFile(**gf_kwargs,
sequencing_experiment_id=seq_exp.kf_id)
db.session.add(gf)
samp.genomic_files.append(gf)
samp.diagnoses.append(diag)
db.session.flush()
rg = ReadGroup(lane_number=4, flow_cell='FL0123')
rg.genomic_files.append(gf)
ct = CavaticaTask(name='task_{}'.format(i))
ct.genomic_files.append(gf)
ca.cavatica_tasks.append(ct)
# Family relationships
for participant1, participant2 in iterate_pairwise(participants):
gender = participant1.gender
rel = 'mother'
if gender == 'male':
rel = 'father'
r = FamilyRelationship(participant1=participant1,
participant2=participant2,
participant1_to_participant2_relation=rel)
db.session.add(r)
db.session.commit()