def _make_family(self, external_id='FAM01'):
"""
Make a family with two participants and a given external id
"""
s = Study(external_id='phs001')
p1 = Participant(external_id="CASE01", is_proband=False)
p2 = Participant(external_id="CASE02", is_proband=False)
s.participants.extend([p1, p2])
f = Family(external_id=external_id, family_type='Duo+')
f.participants.extend([p1, p2])
db.session.add(s)
db.session.commit()
return f
def _create_entities(self):
# Create study
study = Study(external_id='phs001')
# Create participant
p = Participant(external_id='p1',
is_proband=True, study=study)
# Create sequencing_center
sc = SequencingCenter.query.filter_by(name="Baylor").one_or_none()
if sc is None:
sc = SequencingCenter(name="Baylor")
db.session.add(sc)
db.session.commit()
# Create sequencing experiments
se1 = SequencingExperiment(**self._make_seq_exp('se1'),
sequencing_center_id=sc.kf_id)
se2 = SequencingExperiment(**self._make_seq_exp('se2'),
sequencing_center_id=sc.kf_id)
# Create biospecimen
bs = Biospecimen(external_sample_id='bio1', analyte_type='dna',
participant_id=p.kf_id,
sequencing_center_id=sc.kf_id)
# Create genomic files
gfs = []
for i in range(4):
kwargs = {
'file_name': 'file_{}'.format(i),
'data_type': 'submitted aligned read',
'file_format': '.cram',
'urls': ['s3://file_{}'.format(i)],
'hashes': {'md5': str(uuid.uuid4())},
'controlled_access': True,
'is_harmonized': True,
'reference_genome': 'Test01'
}
gf = GenomicFile(**kwargs,
sequencing_experiment_id=se1.kf_id)
if i % 2:
se1.genomic_files.append(gf)
else:
se2.genomic_files.append(gf)
gfs.append(gf)
bs.genomic_files = gfs
p.biospecimens = [bs]
db.session.add(p)
db.session.commit()
def _create_participants_and_studies(self, total):
"""
Creates studies and participants with biospecimens, and diagnoses
"""
# Studies
studies = self._create_studies_investigators()
#
seq_centers = self._create_sequencing_centers()
# Participants
for i in range(total):
diagnoses = self._create_diagnoses(
random.randint(self.min_diagnoses, self.max_diagnoses))
self._get_unique_sites(diagnoses)
biospecimens = self._create_biospecimens(
random.randint(self.min_biospecimens, self.max_biospecimens))
outcomes = self._create_outcomes(
random.randint(self.min_outcomes, self.max_outcomes))
phenotypes = self._create_phenotypes(
random.randint(self.min_phenotypes, self.max_phenotypes))
p = Participant(
external_id='participant_{}'.format(i),
# family_id='family_{}'.format(total % (i + 1)),
is_proband=random.choice(self.is_proband_list),
consent_type=random.choice(self.consent_type_list),
biospecimens=biospecimens,
diagnoses=diagnoses,
outcomes=outcomes,
phenotypes=phenotypes,
study_id=random.choice(studies).kf_id)
db.session.add(p)
f = Family(participants=[p])
db.session.add(f)
db.session.commit()
def _create_save_to_db(self):
"""
Create and save phenotype
Requires creating a participant
Create a phenotype and add it to participant as kwarg
Save participant
"""
# Create study
study = Study(external_id='phs001')
# Create phenotype
kwargs = {
'external_id': 'test_phenotype_0',
'source_text_phenotype': 'Hand Tremor',
'hpo_id_phenotype': 'HP:0002378',
'snomed_id_phenotype': '38033009',
'observed': 'Positive',
'age_at_event_days': 365
}
ph = Phenotype(**kwargs)
# Create and save participant with phenotype
participant_id = 'Test subject 0'
p = Participant(external_id=participant_id,
phenotypes=[ph],
is_proband=True,
study=study)
db.session.add(p)
db.session.commit()
kwargs['participant_id'] = p.kf_id
kwargs['kf_id'] = ph.kf_id
return kwargs
def create_participant_biospecimen(self):
"""
create a participant and biospecimen and save to db
returns participant_id and biospecimen_id
"""
participant_id = "Test_Subject_0"
sample_id = "Test_Sample_0"
aliquot_id = "Test_aliquot_0"
data = self._make_biospecimen(external_sample_id=sample_id,
external_aliquot_id=aliquot_id)
ids = self.create_seqexp()
biospecimen_0 = Biospecimen(**data,
sequencing_center_id=ids
['sequencing_center_id'])
participant_0 = Participant(
external_id=participant_id,
is_proband=True,
biospecimens=[biospecimen_0])
study = Study(external_id='phs001')
study.participants.append(participant_0)
db.session.add(study)
db.session.commit()
return participant_id, sample_id, aliquot_id
def test_create_and_find_biospecimen(self):
"""
Test creation of biospecimen
"""
study = Study(external_id='phs001')
db.session.add(study)
db.session.commit()
dt = datetime.now()
participant_id = "Test_Subject_0"
# creating participant
p = Participant(external_id=participant_id, is_proband=True,
study_id=study.kf_id)
db.session.add(p)
db.session.commit()
# Creating Biospecimen
sample_id = "Test_Sample_0"
aliquot_id = "Test_Aliquot_0"
data = self._make_biospecimen(external_sample_id=sample_id,
external_aliquot_id=aliquot_id)
ids = self.create_seqexp()
s = Biospecimen(**data, participant_id=p.kf_id,
sequencing_center_id=ids['sequencing_center_id'])
db.session.add(s)
db.session.commit()
self.assertEqual(Biospecimen.query.count(), 1)
bs = Biospecimen.query.first()
for key, value in data.items():
self.assertEqual(value, getattr(bs, key))
self.assertGreater(bs.created_at, dt)
self.assertGreater(bs.modified_at, dt)
self.assertIs(type(uuid.UUID(bs.uuid)), uuid.UUID)
def _create_participants_and_dependents(self):
"""
Create participant with required entities
"""
# Create study
study = Study(external_id='phs001')
names = ['Fred', 'Wilma', 'Pebbles', 'Dino']
proband = [True, False]
participants = []
for i, _name in enumerate(names):
# Participants
p = Participant(external_id=_name,
is_proband=random.choice(proband),
study=study)
db.session.add(p)
db.session.commit()
# Sequencing center
sc = SequencingCenter.query.filter_by(name="Baylor").one_or_none()
if sc is None:
sc = SequencingCenter(name="Baylor")
db.session.add(sc)
db.session.commit()
# SequencingExperiment
se = self._create_experiment('se_{}'.format(i),
sequencing_center_id=sc.kf_id)
# Biospecimen
s = self._create_biospecimen('s_{}'.format(i),
sequencing_center_id=sc.kf_id,
participant_id=p.kf_id)
# Input GF
gf_in = self._create_genomic_file(
'gf_{}_in'.format(i),
sequencing_experiment_id=se.kf_id)
# Output GF
gf_out = self._create_genomic_file(
'gf_{}_out'.format(i),
data_type='aligned read',
sequencing_experiment_id=se.kf_id)
s.genomic_files = [gf_in, gf_out]
p.biospecimens = [s]
participants.append(p)
return participants
def _create_all_entities(self):
"""
Create 2 studies with genomic files and read groups
"""
sc = SequencingCenter(name='sc')
studies = []
ses = {}
gfs = {}
for j in range(2):
s = Study(external_id='s{}'.format(j))
p = Participant(external_id='p{}'.format(j))
s.participants.append(p)
study_gfs = gfs.setdefault('study{}'.format(j), [])
for i in range(3):
b = Biospecimen(external_sample_id='b{}'.format(i),
analyte_type='DNA',
sequencing_center=sc,
participant=p)
gf = GenomicFile(
external_id='study{}-gf{}'.format(j, i),
urls=['s3://mybucket/key'],
hashes={'md5': 'd418219b883fce3a085b1b7f38b01e37'})
study_gfs.append(gf)
b.genomic_files.append(gf)
study_ses = ses.setdefault('study{}'.format(j), [])
dt = datetime.now()
kwargs = {
'experiment_date': str(dt.replace(tzinfo=tz.tzutc())),
'experiment_strategy': 'WXS',
'library_name': 'Test_library_name_1',
'library_strand': 'Unstranded',
'is_paired_end': False,
'platform': 'Illumina',
'instrument_model': '454 GS FLX Titanium',
'max_insert_size': 600,
'mean_insert_size': 500,
'mean_depth': 40,
'total_reads': 800,
'mean_read_length': 200
}
se0 = SequencingExperiment(**kwargs,
sequencing_center=sc,
external_id='study{}-se0'.format(j))
se0.genomic_files.extend(study_gfs[0:2])
se1 = SequencingExperiment(**kwargs,
sequencing_center=sc,
external_id='study{}-se1'.format(j))
se1.genomic_files.extend([study_gfs[0], study_gfs[-1]])
study_ses.extend([se0, se1])
studies.append(s)
db.session.add_all(studies)
db.session.commit()
return ses, gfs, studies
def test_modified_at(self):
"""
Test that modified_at is updated when model is updated
"""
s = Study(external_id='phs001')
p = Participant(study=s, external_id='test01', is_proband=True)
db.session.add(p)
db.session.commit()
diff = (p.modified_at - p.created_at)
assert diff < timedelta(seconds=0.01)
time.sleep(0.5)
p.external_id = 'test02'
db.session.add(s)
db.session.commit()
diff = (p.modified_at - p.created_at)
assert diff > timedelta(seconds=0.50)
def _create_entities(self):
"""
Create participant with required entities
"""
# Sequencing center
sc = SequencingCenter.query.filter_by(name="Baylor").one_or_none()
if sc is None:
sc = SequencingCenter(name="Baylor")
db.session.add(sc)
db.session.commit()
# Create study
study = Study(external_id='phs001')
# Participants
p = Participant(external_id='p0', is_proband=True, study=study)
# Biospecimen
bs = Biospecimen(analyte_type='dna',
sequencing_center=sc,
participant=p)
# SequencingExperiment
data = {
'external_id': 'se',
'experiment_strategy': 'wgs',
'is_paired_end': True,
'platform': 'platform',
'sequencing_center': sc
}
se = SequencingExperiment(**data)
# Genomic Files
genomic_files = []
for i in range(4):
data = {
'file_name': 'gf_{}'.format(i),
'data_type': 'submitted aligned read',
'file_format': '.cram',
'urls': ['s3://file_{}'.format(i)],
'hashes': {
'md5': str(uuid.uuid4())
},
'is_harmonized': True if i % 2 else False
}
gf = GenomicFile(**data)
bs.genomic_files.append(gf)
se.genomic_files.append(gf)
genomic_files.append(gf)
ct = self._create_cavatica_task('ct1')
db.session.add(ct)
db.session.add(study)
db.session.commit()
def _create_all_entities():
"""
Create 2 studies with genomic files and read groups
"""
sc = SequencingCenter(name='sc')
studies = []
ses = {}
rgs = {}
gfs = {}
for j in range(2):
s = Study(external_id='s{}'.format(j))
p = Participant(external_id='p{}'.format(j))
s.participants.append(p)
study_gfs = gfs.setdefault('study{}'.format(j), [])
for i in range(3):
b = Biospecimen(external_sample_id='b{}'.format(i),
analyte_type='DNA',
sequencing_center=sc,
participant=p)
gf = GenomicFile(
external_id='study{}-gf{}'.format(j, i),
urls=['s3://mybucket/key', 'https://gen3.something.com/did'],
hashes={'md5': 'd418219b883fce3a085b1b7f38b01e37'})
study_gfs.append(gf)
b.genomic_files.append(gf)
study_rgs = rgs.setdefault('study{}'.format(j), [])
rg0 = ReadGroup(external_id='study{}-rg0'.format(j))
rg0.genomic_files.extend(study_gfs[0:2])
rg1 = ReadGroup(external_id='study{}-rg1'.format(j))
rg1.genomic_files.extend([study_gfs[0], study_gfs[-1]])
study_ses = ses.setdefault('study{}'.format(j), [])
se0 = SequencingExperiment(external_id='study{}-se0'.format(j),
experiment_strategy='WGS',
is_paired_end=True,
platform='platform',
sequencing_center=sc)
se0.genomic_files.extend(study_gfs[0:2])
se1 = SequencingExperiment(external_id='study{}-se1'.format(j),
experiment_strategy='WGS',
is_paired_end=True,
platform='platform',
sequencing_center=sc)
se1.genomic_files.extend([study_gfs[0], study_gfs[-1]])
study_rgs.extend([rg0, rg1])
study_ses.extend([se0, se1])
studies.append(s)
db.session.add_all(studies)
db.session.commit()
return ses, rgs, gfs, studies
def _create_save_to_db(self):
"""
Create and save family_relationship
Requires creating a participant
Create a family_relationship and add it to participant as kwarg
Save participant
"""
# Create study
study = Study(external_id='phs001')
# Create participants
p1 = Participant(external_id='Fred', is_proband=False)
p2 = Participant(external_id='Wilma', is_proband=False)
p3 = Participant(external_id='Pebbles', is_proband=True)
p4 = Participant(external_id='Dino', is_proband=True)
study.participants.extend([p1, p2, p3, p4])
db.session.add(study)
db.session.commit()
# Create family_relationship
kwargs = {
'participant1_id': p1.kf_id,
'participant2_id': p3.kf_id,
'participant1_to_participant2_relation': 'father',
'source_text_notes': 'Notes 1'
}
fr = FamilyRelationship(**kwargs)
db.session.add(fr)
db.session.commit()
kwargs['kf_id'] = fr.kf_id
kwargs['participant2_to_participant1_relation'] = \
fr.participant2_to_participant1_relation
fr.external_id = str(fr)
db.session.commit()
return p1, p2, p3, p4, study, kwargs
def _create_save_to_db(self):
"""
Create and save diagnosis
Requires creating a participant
Create a diagnosis and add it to participant as kwarg
Save participant
"""
# Create study
study = Study(external_id='phs001')
# Create diagnosis
kwargs = {
'external_id': 'd1',
'source_text_diagnosis': 'flu',
'diagnosis_category': 'Cancer',
'source_text_tumor_location': 'Brain',
'age_at_event_days': 365,
'mondo_id_diagnosis': 'DOID:8469',
'icd_id_diagnosis': 'J10.01',
'uberon_id_tumor_location': 'UBERON:0000955',
'spatial_descriptor': 'left side'
}
d = Diagnosis(**kwargs)
# Create and save participant with diagnosis
participant_id = 'Test subject 0'
p = Participant(external_id=participant_id,
diagnoses=[d],
is_proband=True,
study=study)
db.session.add(p)
db.session.commit()
# Create sequencing center
s = SequencingCenter(name='washu')
db.session.add(s)
db.session.commit()
# Create biospecimen
b = Biospecimen(analyte_type='DNA',
sequencing_center_id=s.kf_id,
participant=p)
db.session.add(s)
db.session.add(b)
db.session.commit()
kwargs['participant_id'] = p.kf_id
kwargs['kf_id'] = d.kf_id
return kwargs
def _create_save_dependents(self):
"""
Create and save all genomic file dependent entities to db
Dependent entities: participant, biospecimens
"""
# Create study
study = Study(external_id='phs001')
# Create participant
p = Participant(external_id='p1',
biospecimens=self._create_biospecimens(),
is_proband=True, study=study)
db.session.add(p)
db.session.commit()
def test_delete_orphans(self):
"""
Test that orphaned alias groups are deleted
Orphans are alias groups with 0 particpants
"""
# Create alias group
data = self._create_save_to_db()
# Create another alias group
study = Study.query.first()
p6 = Participant(external_id='p6',
is_proband=True,
study_id=study.kf_id)
p7 = Participant(external_id='p7',
is_proband=True,
study_id=study.kf_id)
p6.add_alias(p7)
db.session.add_all([p6, p7])
db.session.commit()
self.assertEqual(2, AliasGroup.query.count())
# Make orphan
groups = AliasGroup.query.all()
for p in groups[0].participants:
db.session.delete(p)
db.session.commit()
# Check that the orphan was deleted and other ag was unaffected
self.assertEqual(1, AliasGroup.query.count())
self.assertEqual(len(groups[1].participants),
len(AliasGroup.query.first().participants))
# Check that ag w at least 1 particpant does not get deleted
db.session.delete(groups[1].participants[0])
db.session.commit()
self.assertEqual(1, AliasGroup.query.count())
def test_link_biospecimen_diagnosis(self):
"""
Test Deleting one of the biospecimens
"""
# create a participant with a biospecimen
(participant_id,
sample_id,
aliquot_id) = self.create_participant_biospecimen()
p = Participant.query.first()
# Create diagnosis
kwargs = {
'external_id': 'id_1',
'source_text_diagnosis': 'diagnosis_1',
'age_at_event_days': 365,
'diagnosis_category': 'cancer',
'source_text_tumor_location': 'Brain',
'mondo_id_diagnosis': 'DOID:8469',
'uberon_id_tumor_location': 'UBERON:0000955',
'icd_id_diagnosis': 'J10.01',
'spatial_descriptor': 'left side',
'participant_id': p.kf_id
}
dg = Diagnosis(**kwargs)
db.session.add(dg)
biospecimen = Biospecimen.query.first()
# create link btn bs and ds
bs_ds = BiospecimenDiagnosis(biospecimen_id=biospecimen.kf_id,
diagnosis_id=dg.kf_id)
db.session.add(bs_ds)
db.session.commit()
self.assertEqual(BiospecimenDiagnosis.query.count(), 1)
self.assertEqual(bs_ds.biospecimen_id, biospecimen.kf_id)
self.assertEqual(bs_ds.diagnosis_id, dg.kf_id)
s = Study(external_id="study")
sc = SequencingCenter.query.first()
p1 = Participant(external_id='p1', study=s)
b1 = Biospecimen(analyte_type='RNA', participant=p1,
sequencing_center_id=sc.kf_id)
db.session.add(s)
db.session.commit()
# Participant 1 - Link their biop b1 to Participant 0 diagnosis d0
bd1 = BiospecimenDiagnosis(biospecimen_id=b1.kf_id,
diagnosis_id=dg.kf_id)
db.session.add(bd1)
with self.assertRaises(DatabaseValidationError):
db.session.commit()
db.session.rollback()
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 _create_participant(self, external_id='Test_Participant_0'):
"""
Create participant with external id
"""
s = Study(external_id='phs001')
data = {
'external_id': external_id,
'is_proband': False,
'race': 'asian',
'ethnicity': 'not hispanic',
'gender': 'female',
'affected_status': False,
'diagnosis_category': 'Cancer'
}
p = Participant(**data)
s.participants.append(p)
db.session.add(s)
db.session.commit()
return p
def _create_save_participants(self, n=5):
"""
Create participants
"""
s = Study(external_id='phs001')
particpant_data = {}
for i in range(n):
k = 'p{}'.format(i)
particpant_data[k] = {
'external_id': k,
'is_proband': random.choice([True, False])
}
pt = Participant(**particpant_data[k])
particpant_data[k]['obj'] = pt
s.participants.append(pt)
db.session.add(s)
db.session.commit()
return particpant_data
def _create_phenotypes(self):
"""
Create phenotypes and required entities
"""
# Create Study
study = Study(external_id='phs001')
# Create two phenotypes
pheno = ['test phenotype 1', 'test phenotype 2']
ph1 = Phenotype(source_text_phenotype=pheno[0],
external_id='test_phenotype_0')
ph2 = Phenotype(source_text_phenotype=pheno[1],
external_id='test_phenotype_0')
p = Participant(external_id='p1', is_proband=True, study=study)
# Add to participant and save
p.phenotypes.extend([ph1, ph2])
db.session.add(p)
db.session.commit()
return [ph1, ph2], p, pheno
def _create_all_entities(self):
"""
Create 2 studies with genomic files and read groups
"""
sc = SequencingCenter(name='sc')
studies = []
rgs = {}
gfs = {}
for j in range(2):
s = Study(external_id='s{}'.format(j))
p = Participant(external_id='p{}'.format(j))
s.participants.append(p)
study_gfs = gfs.setdefault('study{}'.format(j), [])
for i in range(3):
b = Biospecimen(external_sample_id='b{}'.format(i),
analyte_type='DNA',
sequencing_center=sc,
participant=p)
gf = GenomicFile(
external_id='study{}-gf{}'.format(j, i),
urls=['s3://mybucket/key'],
hashes={'md5': 'd418219b883fce3a085b1b7f38b01e37'})
study_gfs.append(gf)
b.genomic_files.append(gf)
study_rgs = rgs.setdefault('study{}'.format(j), [])
rg0 = ReadGroup(external_id='study{}-rg0'.format(j))
rg0.genomic_files.extend(study_gfs[0:2])
rg1 = ReadGroup(external_id='study{}-rg1'.format(j))
rg1.genomic_files.extend([study_gfs[0], study_gfs[-1]])
study_rgs.extend([rg0, rg1])
studies.append(s)
db.session.add_all(studies)
db.session.commit()
return rgs, gfs, studies
def test_create_and_find(self):
"""
Test create phenotype
"""
dt = datetime.now()
# 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()
kwarg_dict = {}
# Create phenotypes
for i in range(2):
data = {
'external_id': 'test_phenotype_{}'.format(i),
'source_text_phenotype': 'test phenotype_{}'.format(i),
'hpo_id_phenotype': 'HP:0000118',
'snomed_id_phenotype': '38033009',
'age_at_event_days': 120,
'participant_id': p.kf_id
}
ph = Phenotype(**data)
kwarg_dict[ph.external_id] = data
db.session.add(ph)
db.session.commit()
self.assertEqual(Phenotype.query.count(), 2)
for k, kwargs in kwarg_dict.items():
ph = Phenotype.query.filter_by(external_id=k).one()
for key, value in kwargs.items():
self.assertEqual(value, getattr(ph, key))
self.assertGreater(ph.created_at, dt)
self.assertGreater(ph.modified_at, dt)
self.assertIs(type(uuid.UUID(ph.uuid)), uuid.UUID)
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_post(self):
"""
Test create a new phenotype
"""
# Create study
study = Study(external_id='phs001')
# Create a participant
p = Participant(external_id='Test subject 0',
is_proband=True,
study=study)
db.session.add(p)
db.session.commit()
# Create phenotype data
kwargs = {
'external_id': 'test_phenotype_0',
'source_text_phenotype': 'Hand tremor',
'age_at_event_days': 365,
'hpo_id_phenotype': 'HP:0002378',
'observed': 'Positive',
'participant_id': p.kf_id
}
# Send get request
response = self.client.post(url_for(PHENOTYPES_LIST_URL),
data=json.dumps(kwargs),
headers=self._api_headers())
# Check response status status_code
self.assertEqual(response.status_code, 201)
# Check response content
response = json.loads(response.data.decode('utf-8'))
phenotype = response['results']
ph = Phenotype.query.get(phenotype.get('kf_id'))
for k, v in kwargs.items():
if k == 'participant_id':
continue
self.assertEqual(phenotype[k], getattr(ph, k))
def test_add_invalid_biospecimen(self):
"""
Test that a diagnosis cannot be linked with a biospecimen if
they refer to different participants
"""
diagnoses, kwarg_dict = self._create_diagnoses()
# Get first participant
st = Study.query.first()
s = SequencingCenter.query.first()
# Create new participant with biospecimen
p1 = Participant(external_id='p1', is_proband=True, study_id=st.kf_id)
b = Biospecimen(analyte_type='DNA',
sequencing_center_id=s.kf_id,
participant=p1)
db.session.add(b)
db.session.commit()
# Try linking
d = Diagnosis.query.first()
b.diagnoses.append(d)
with self.assertRaises(DatabaseValidationError):
db.session.commit()
db.session.rollback()
def _create_diagnoses(self, total=2):
"""
Create diagnoses and other requred entities
"""
# Create study
study = Study(external_id='phs001')
# Create participant
participant_id = 'Test subject 0'
p = Participant(external_id=participant_id,
is_proband=True,
study=study)
# Create sequencing center
s = SequencingCenter(name='washu')
db.session.add(s)
db.session.commit()
# Create biospecimen
b = Biospecimen(analyte_type='DNA',
sequencing_center_id=s.kf_id,
participant=p)
db.session.add(p)
db.session.add(b)
db.session.commit()
# Create diagnoses
diagnoses = []
kwarg_dict = {}
for i in range(total):
d, kwargs = self._create_diagnosis(i, participant_id=p.kf_id)
kwarg_dict[d.external_id] = kwargs
diagnoses.append(d)
db.session.add_all(diagnoses)
db.session.commit()
return diagnoses, kwarg_dict
def test_special_filter_param(self):
"""
Test special filter param participant_id
/family-relationships?participant_id
"""
# Add some family relationships
p1, p2, p3, p4, s1, kwargs = self._create_save_to_db()
r2 = FamilyRelationship(participant1=p1,
participant2=p4,
participant1_to_participant2_relation='father')
r3 = FamilyRelationship(participant1=p2,
participant2=p3,
participant1_to_participant2_relation='mother')
r4 = FamilyRelationship(participant1=p2,
participant2=p4,
participant1_to_participant2_relation='mother')
db.session.add_all([r2, r3, r4])
db.session.commit()
# Case 1 - Participant with no family defined
url = (url_for(FAMILY_RELATIONSHIPS_LIST_URL) +
'?participant_id={}'.format(p3.kf_id))
response = self.client.get(url, headers=self._api_headers())
self.assertEqual(response.status_code, 200)
response = json.loads(response.data.decode("utf-8"))
content = response.get('results')
# Only immediate family relationships returned
self.assertEqual(len(content), 2)
# Test with additional filter parameters
url = (url_for(FAMILY_RELATIONSHIPS_LIST_URL) + '?participant_id={}'
'&study_id={}&participant1_to_participant2_relation={}'.format(
p3.kf_id, s1.kf_id, 'father'))
response = self.client.get(url, headers=self._api_headers())
self.assertEqual(response.status_code, 200)
response = json.loads(response.data.decode("utf-8"))
content = response.get('results')
self.assertEqual(len(content), 1)
# Case 2 - Participant with a family defined
f0 = Family(external_id='phs001-family')
f0.participants.extend([p1, p2, p3, p4])
db.session.add(f0)
db.session.commit()
url = (url_for(FAMILY_RELATIONSHIPS_LIST_URL) +
'?participant_id={}'.format(p3.kf_id))
response = self.client.get(url, headers=self._api_headers())
self.assertEqual(response.status_code, 200)
response = json.loads(response.data.decode("utf-8"))
content = response.get('results')
# All family relationships returned
self.assertEqual(len(content), 4)
# Add another study with a family and relationships
s2 = Study(external_id='phs002')
f2 = Family(external_id='phs002-family')
p_1 = Participant(external_id='Fred_1', is_proband=False)
p_2 = Participant(external_id='Wilma_1', is_proband=False)
p_3 = Participant(external_id='Pebbles_1', is_proband=True)
r_1 = FamilyRelationship(
participant1=p_1,
participant2=p_3,
participant1_to_participant2_relation='father')
r_2 = FamilyRelationship(
participant1=p_2,
participant2=p_3,
participant1_to_participant2_relation='mother')
s2.participants.extend([p_1, p_2, p_3])
f2.participants.extend([p_1, p_2, p_3])
db.session.add(s2)
db.session.add(f2)
db.session.add_all([r_1, r_2])
db.session.commit()
# Should see same results for p3
url = (url_for(FAMILY_RELATIONSHIPS_LIST_URL) +
'?participant_id={}'.format(p3.kf_id))
response = self.client.get(url, headers=self._api_headers())
self.assertEqual(response.status_code, 200)
response = json.loads(response.data.decode("utf-8"))
content = response.get('results')
# All family relationships returned
self.assertEqual(len(content), 4)
def _create_save_to_db(self):
"""
Create and save biospecimen
Requires creating a participant
Create a biospecimen and add it to participant as kwarg
Save participant
"""
dt = datetime.now()
study = Study(external_id='phs001')
db.session.add(study)
db.session.commit()
sc = SequencingCenter.query.filter_by(name="Baylor").one_or_none()
if sc is None:
sc = SequencingCenter(name="Baylor")
db.session.add(sc)
db.session.commit()
se = SequencingExperiment(external_id="Test_seq_ex_o",
experiment_strategy="WGS",
is_paired_end="True",
platform="Test_platform",
sequencing_center_id=sc.kf_id)
db.session.add(se)
db.session.commit()
# Create biospecimen
kwargs = {
'external_sample_id': 's1',
'external_aliquot_id': 'a1',
'source_text_tissue_type': 'Normal',
'composition': 'composition1',
'source_text_anatomical_site': 'Brain',
'age_at_event_days': 365,
'source_text_tumor_descriptor': 'Metastatic',
'shipment_origin': 'CORIELL',
'analyte_type': 'DNA',
'concentration_mg_per_ml': 100,
'volume_ul': 12.67,
'shipment_date': dt,
'spatial_descriptor': 'left side',
'ncit_id_tissue_type': 'Test',
'ncit_id_anatomical_site': 'C12439',
'uberon_id_anatomical_site': 'UBERON:0000955',
'consent_type': 'GRU-IRB',
'dbgap_consent_code': 'phs00000.c1',
'sequencing_center_id': sc.kf_id
}
d = Biospecimen(**kwargs)
# Create and save participant with biospecimen
p = Participant(external_id='Test subject 0',
biospecimens=[d],
is_proband=True,
study_id=study.kf_id)
db.session.add(p)
db.session.commit()
kwargs['participant_id'] = p.kf_id
kwargs['kf_id'] = d.kf_id
return kwargs
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()
def _create_all_entities(self):
"""
Create 2 studies with same content
Content: 3 participants, 4 biospecimens, 4 diagnoses
"""
# Create entities
sc = SequencingCenter.query.filter_by(name='sc').first()
if not sc:
sc = SequencingCenter(name='sc')
studies = []
# Two studies
for j in range(2):
s = Study(external_id='s{}'.format(j))
p0 = Participant(external_id='study{}-p0'.format(j))
p1 = Participant(external_id='study{}-p1'.format(j))
p2 = Participant(external_id='study{}-p2'.format(j))
# Participant 0
# Has 2 Biospecimens
for i in range(2):
b = Biospecimen(external_sample_id='study{}-p0-b{}'.format(
j, i),
analyte_type='DNA',
sequencing_center=sc)
# Biospecimen b0 has 2 diagnoses
if i == 0:
for k in range(2):
d = Diagnosis(
external_id='study{}-p0-d{}'.format(j, k))
p0.diagnoses.append(d)
# Biospecimen b1 has 1 diagnosis
else:
d = Diagnosis(
external_id='study{}-p0-d{}'.format(j, k + 1))
p0.diagnoses.append(d)
p0.biospecimens.append(b)
# Participant 1
# Has 1 biospecimen, 1 diagnosis
b = Biospecimen(external_sample_id='study{}-p1-b0'.format(j),
analyte_type='DNA',
sequencing_center=sc)
d = Diagnosis(external_id='study{}-p1-d0'.format(j))
p1.biospecimens.append(b)
p1.diagnoses.append(d)
# Participant 2
# Has 1 biospecimen
b = Biospecimen(external_sample_id='study{}-p2-b0'.format(j),
analyte_type='DNA',
sequencing_center=sc)
p2.biospecimens.append(b)
s.participants.extend([p0, p1, p2])
studies.append(s)
db.session.add_all(studies)
db.session.commit()
# Create links between bios and diags
bs_dgs = []
# Participant 0
p0 = studies[0].participants[0]
# b0-d0
bs_dgs.append(
BiospecimenDiagnosis(biospecimen_id=p0.biospecimens[0].kf_id,
diagnosis_id=p0.diagnoses[0].kf_id))
# b0-d1
bs_dgs.append(
BiospecimenDiagnosis(biospecimen_id=p0.biospecimens[0].kf_id,
diagnosis_id=p0.diagnoses[1].kf_id))
# b1-d2
bs_dgs.append(
BiospecimenDiagnosis(biospecimen_id=p0.biospecimens[1].kf_id,
diagnosis_id=p0.diagnoses[2].kf_id))
# b0-d2
bs_dgs.append(
BiospecimenDiagnosis(biospecimen_id=p0.biospecimens[0].kf_id,
diagnosis_id=p0.diagnoses[2].kf_id))
# Participant 1
p1 = studies[0].participants[1]
# b0-d0
bs_dgs.append(
BiospecimenDiagnosis(biospecimen_id=p1.biospecimens[0].kf_id,
diagnosis_id=p1.diagnoses[0].kf_id))
db.session.add_all(bs_dgs)
db.session.commit()