def _create_outcomes(self):
"""
Create outcome and required entities
"""
# Create study
study = Study(external_id='phs001')
# Create two outcomes
oc = ['Deceased', 'Alive']
o1 = Outcome(vital_status=oc[0])
o2 = Outcome(vital_status=oc[1])
p = Participant(external_id='p1', is_proband=True, study=study)
# Add to participant and save
p.outcomes.extend([o1, o2])
db.session.add(p)
db.session.commit()
return [o1, o2], p
def test_create(self):
"""
Test create outcome
"""
# Create study
study = Study(external_id='phs001')
# Create and save participant
participant_id = 'Test subject 0'
p = Participant(external_id=participant_id,
is_proband=True,
study=study)
db.session.add(p)
db.session.commit()
# Create outcomes
data = {
'external_id': 'test_0',
'vital_status': 'Alive',
'disease_related': False,
'age_at_event_days': 120,
'participant_id': p.kf_id
}
dt = datetime.now()
o1 = Outcome(**data)
db.session.add(o1)
data['vital_status'] = 'Deceased'
data['disease_related'] = 'True'
o2 = Outcome(**data)
db.session.add(o2)
db.session.commit()
self.assertEqual(Outcome.query.count(), 2)
new_outcome = Outcome.query.all()[1]
self.assertGreater(new_outcome.created_at, dt)
self.assertGreater(new_outcome.modified_at, dt)
self.assertIs(type(uuid.UUID(new_outcome.uuid)), uuid.UUID)
self.assertEqual(new_outcome.vital_status, data['vital_status'])
self.assertEqual(new_outcome.disease_related, data['disease_related'])
def test_foreign_key_constraint(self):
"""
Test that a outcome cannot be created without an existing
reference Participant. This checks foreign key constraint
"""
# Create outcome
data = {
'vital_status': 'Alive',
'participant_id': '' # empty blank foreign key
}
o = Outcome(**data)
# Add to db
self.assertRaises(IntegrityError, db.session.add(o))
def test_not_null_constraint(self):
"""
Test that a outcome cannot be created without required
parameters such as participant_id
"""
# Create outcome
data = {
'vital_status': 'Alive',
# non-existent required param: participant_id
}
o = Outcome(**data)
# Add to db
self.assertRaises(IntegrityError, db.session.add(o))
def _create_outcomes(self, total):
"""
creates outcomes
"""
outcomes_list = []
for i in range(total):
# Fill disaese_related only if person is dead
vs = random.choice(self.vital_status_list)
if vs == 'Dead':
dr = random.choice(self.disease_related_list)
data = {
'vital_status': vs,
'disease_related': dr,
'age_at_event_days': random.randint(0, 32872)
}
outcomes_list.append(Outcome(**data))
break
else:
data = {
'vital_status': vs,
'age_at_event_days': random.randint(0, 32872)
}
outcomes_list.append(Outcome(**data))
return outcomes_list
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()