def delete_genes_test_data():
genes = SESSION.query(Gene).all()
# each Gene needs to be deleted individually in order to delete
# the associated records from table 'gene_ontology_map';
# bulk deleting all Gene instances doesn't delete the records in 'gene_ontology_map'
for gene in genes:
SESSION.delete(gene)
def delete_test_ontology_terms_data():
on_terms = SESSION.query(OntologyTerm).all()
# each OntologyTerm needs to be deleted individually in order to delete
# the associated records from table 'on_pairs';
# bulk deleting all OntologyTerm instances doesn't delete the records in 'on_pairs'
for term in on_terms:
SESSION.delete(term)
def test_get_cytogenetic_band_neg(self):
""" NEGATIVE CASE: Test to check non-existing taxon ID - Zebrafish - 7955 """
neg_taxonid = 7955
cytogenetic_band = SESSION.query(CytogeneticBand) \
.filter(CytogeneticBand.taxon_id == neg_taxonid) \
.first()
self.assertIsNone(cytogenetic_band)
def test_get_all_synteny_blocks(self):
""" Test getting back all SyntenyBlock entries """
blocks = SESSION.query(SyntenicBlock).all()
self.assertTrue(len(blocks) == len(SYNTENY_BLOCKS_DATA))
for i, block in enumerate(blocks):
serialized = marshal(block, BLOCKS_SCHEMA)
self.assertTrue(serialized['id']
in SYNTENY_BLOCKS_DATA[i][9])
def test_get_all_loci(self):
"""
Positive case: test getting back all QTL records.
"""
loci = SESSION.query(Feature).all()
self.assertTrue(len(loci), len(QTLS_DATA))
for i, locus in enumerate(loci):
serialized = marshal(locus, QTLS_SCHEMA)
self.assertTrue(serialized['id'] in QTLS_DATA[i][11])
def get_genes_by_species(species_id):
"""
Function that queries the database and returns a list of Gene objects for a specific species.
:param species_id: NCBI species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:return: genes - a list of Gene objects or an empty list
"""
query = SESSION.query(Gene).filter_by(taxon_id=species_id)
genes = query.all()
return genes
def get_all_genes():
"""
Function that queries the database and returns a list of all Gene objects available.
:return: genes - a list of Gene objects or an empty list
"""
query = SESSION.query(Gene)
genes = query.all()
return genes
def get_all_snps():
"""
Function that queries the database and returns a list of all available 'SNP Variant' objects.
:return: snps - a list of 'SnpVariant' objects or an empty list (if none exist)
"""
query = SESSION.query(SnpVariant)
snps = query.all()
return snps
def get_all_bands():
"""
Function that queries the database and returns a list of all 'Cytogenetic Band' objects available.
:return: genes - a list of 'Cytogenetic Band' objects or an empty list
"""
query = SESSION.query(CytogeneticBand)
bands = query.all()
return bands
def test_get_synteny_block_by_species_ids(self):
""" Test getting back entries by reference and comparison species IDs """
ref_taxonid = SYNTENY_BLOCKS_DATA[0][0]
comp_taxonid = SYNTENY_BLOCKS_DATA[0][4]
block = SESSION.query(SyntenicBlock)\
.filter(and_(SyntenicBlock.ref_taxonid == ref_taxonid,
SyntenicBlock.comp_taxonid == comp_taxonid))\
.first()
self.assertIsNotNone(block)
self.assertEqual(block.id, SYNTENY_BLOCKS_DATA[0][9])
def test_get_cytogenetic_band(self):
""" POSITIVE CASE: Test getting back all cytogenetic bands. """
cytogenetic_band = SESSION.query(CytogeneticBand).all()
self.assertTrue(len(cytogenetic_band) == len(CYTOGENETIC_BAND_DATA))
for i, band in enumerate(cytogenetic_band):
logging.debug(i)
serialized = marshal(band, CYTOGENETIC_BAND_SCHEMA)
self.assertTrue(
serialized['location'] in CYTOGENETIC_BAND_DATA[i][7])
def get_homologs_by_species_ids_and_reference_chromosome(
ref_taxonid, comp_taxonid, chromosome):
"""
:param ref_taxonid: species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:param comp_taxonid: species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:param chromosome: reference species chromosome ID
:return: homologs - a list of Gene objects or an empty list
"""
# select all reference species genes,
# located on the specified chromosome
genes_list = SESSION.query(Gene) \
.filter(and_(Gene.chr == chromosome,
Gene.taxon_id == ref_taxonid)) \
.all()
# iterate through the gene list and identify all
# homologs that belong to the comparison species
homologs_set = set()
for g in genes_list:
for h in g.homologs:
if h.taxon_id == comp_taxonid:
homologs_set.add(h.id)
# the maximum number of host parameters in a single
# SQL statement in SQLite is 999. Chunk the data so that
# the request does not result in 'sqlite.OperationalError: too many SQL variables'
sqlite_max_variable_num = 999
# convert the set to list (since lists can be indexed)
homologs_list = list(homologs_set)
chunks = [
homologs_list[x:x + sqlite_max_variable_num - 1]
for x in range(0, len(homologs_list), sqlite_max_variable_num - 1)
]
# homologs list
homologs = []
for chunk in chunks:
# select all (homolog) genes: these are all comparison species
# genes, which are located on various chromosomes and are homologs
# to all reference species genes, located on the specified chromosome
# TODO: [1/3/2020 gik] consider using the Homolog table for the comparison genes information, rather than Gene
# TODO: [1/3/2020 gik] it is possible that the Gene table doesn't have entry for the comparison gene(s)
query = SESSION \
.query(Gene) \
.filter(and_(Gene.taxon_id == comp_taxonid, Gene.id.in_(chunk)))\
.order_by(Gene.id)
homologs.extend(query.all())
return homologs
def test_query_nonexistent_reference_species(self):
""" Test that getting data for non-existent reference species has no result """
# 7227 is Drosophila melanogaster's NSBI species taxonomy ID
ref_taxonid = 7227
comp_taxonid = SYNTENY_BLOCKS_DATA[4][4]
block = SESSION.query(SyntenicBlock) \
.filter(and_(SyntenicBlock.ref_taxonid == ref_taxonid,
SyntenicBlock.comp_taxonid == comp_taxonid)) \
.first()
self.assertIsNone(block)
def get_snps_by_species_and_trait(species_id, trait_id):
"""
Function that returns a list of 'SNP Variant' objects for the specified species and trait.
:param species_id: NCBI assigned species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:param trait_id: SNP source (such as EBI) assigned trait ID (e.g. 0001360 for Type 2 Diabetes), etc.
:return: snps - a list of 'SnpVariant' objects or an empty list (if none exist)
"""
query = SESSION.query(SnpVariant).filter_by(taxon_id=species_id,
trait_id=trait_id)
snps = query.all()
return snps
def get_blocks_by_species_ids(ref_taxonid, comp_taxonid):
"""
Queries the database and returns a list of SyntenyBlock objects for specific reference and comparison species.
:param ref_taxonid: NCBI species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:param comp_taxonid: NCBI species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:return: blocks - a list of SyntenyBlock objects or an empty list
"""
query = SESSION.query(SyntenicBlock) \
.filter(and_(SyntenicBlock.ref_taxonid == ref_taxonid,
SyntenicBlock.comp_taxonid == comp_taxonid))
blocks = query.all()
return blocks
def test_get_synteny_blocks_by_taxon_ids_chromosome(self):
""" Test getting back entries by reference and comparison species IDs, and a reference species chromosome """
ref_taxonid = SYNTENY_BLOCKS_DATA[4][0]
ref_chromosome = SYNTENY_BLOCKS_DATA[4][1]
comp_taxonid = SYNTENY_BLOCKS_DATA[4][4]
block = SESSION.query(SyntenicBlock) \
.filter(and_(SyntenicBlock.ref_taxonid == ref_taxonid,
SyntenicBlock.comp_taxonid == comp_taxonid,
SyntenicBlock.ref_chr == ref_chromosome)) \
.first()
self.assertIsNotNone(block)
self.assertEqual(block.id, SYNTENY_BLOCKS_DATA[4][9])
def get_bands_by_species_and_chromosome(species_id, chromosome):
"""
Function that queries the database and returns a list of 'Cytogenetic Band' objects for a specific species and chromosome.
:param species_id: NCBI species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:param chromosome: reference species chromosome ID
:return: genes - a list of 'Cytogenetic Band' objects or an empty list
"""
query = SESSION.query(CytogeneticBand).filter_by(
taxon_id=species_id,
chr=chromosome
)
bands = query.all()
return bands
def test_get_homologs_by_species_ids_and_reference_chromosome(self):
""" Test getting back homologs by reference and comparison species IDs, and by reference species chromosome. """
ref_taxonid = GENES_DATA[7][10][0][2]
ref_chromosome = GENES_DATA[7][10][0][3]
comp_taxonid = GENES_DATA[7][10][0][8]
homolog = SESSION.query(Homolog)\
.filter(and_(Homolog.ref_taxon_id == ref_taxonid,
Homolog.ref_seq_id == ref_chromosome,
Homolog.taxon_id == comp_taxonid))\
.first()
self.assertIsNotNone(homolog)
self.assertEqual(homolog.id, GENES_DATA[0][0])
def test_query_nonexistent_reference_species(self):
""" Test that getting data for non-existent reference species has no result """
# 7227 is Drosophila melanogaster's NCBI species taxonomy ID
ref_taxonid = 7227
ref_chromosome = GENES_DATA[7][10][0][3]
comp_taxonid = GENES_DATA[7][10][0][8]
homolog = SESSION.query(Homolog) \
.filter(and_(Homolog.ref_taxon_id == ref_taxonid,
Homolog.ref_seq_id == ref_chromosome,
Homolog.taxon_id == comp_taxonid)) \
.first()
self.assertIsNone(homolog)
def test_get_loci_when_valid_species_id(self):
"""
Positive case: test getting back all QTL records for a specific valid species.
"""
species_id = QTLS_DATA[0][0]
expected_loci_ids = [
locus[11] for locus in QTLS_DATA if locus[0] == species_id
]
loci = SESSION.query(Feature)\
.filter(and_(Feature.taxon_id == species_id, Feature.type == 'QTL')).all()
self.assertIsNotNone(loci)
for locus in loci:
serialized = marshal(locus, QTLS_SCHEMA)
self.assertTrue(serialized["id"] in expected_loci_ids)
def get_genes_by_species_chromosome_position(species_id, chromosome,
block_start, block_end):
"""
Returns a list of Gene objects selected based on specific species, chromosome, and range on the chromosome.
:param species_id: NCBI species ID, such as 9606 (H. sapiens), 10090 (M. musculus), etc.
:param chromosome: species chromosome ID
:block_start: synteny block start position
:block_end: synteny block end position
:return: genes - a list of Gene objects or an empty list
"""
query = SESSION.query(Gene).filter(
and_(Gene.taxon_id == species_id, Gene.chr == chromosome,
Gene.start <= block_end, Gene.end >= block_start))
genes = query.all()
return genes
def test_get_metadata_by_ontology_term_id(self):
""" positive case: test selecting term and it's descendants based on specific term id - such as GO:0046983 """
# ontology term descendant its from test data
expected_descendant_ids = [];
for term in ONTOLOGY_TERMS_DATA:
if term[0] == 'GO:0046983':
# check and add descendants:
# those would have been entered in the DB by the model
for descendant in term[5]:
expected_descendant_ids.append(descendant[0])
break
term = SESSION.query(OntologyTerm) \
.filter_by(id='GO:0046983').all()
serialized = marshal(term, ontologies_controller.ONT_TERM_METADATA_SCHEMA)
self.assertEqual(len(serialized), 1)
self.assertEqual(len(serialized[0]["descendants"]), len(expected_descendant_ids))
for d in serialized[0]["descendants"]:
self.assertTrue(d["id"] in expected_descendant_ids)
def test_get_terms_by_ontology_prefix(self):
""" positive case: test selecting ontology terms with specific prefix - such as GO (Gene Ontology). """
# GO ontology term ids from test data
expected_term_ids = []
for term in ONTOLOGY_TERMS_DATA:
if term[0].startswith('GO:'):
expected_term_ids.append(term[0])
# check and add descendants:
# those would have been entered in the DB by the model
for d in term[5]:
expected_term_ids.append(d[0])
# remove duplicated ids: test data shouldn't have any, but just in case
expected_term_ids = list(dict.fromkeys(expected_term_ids))
terms = SESSION.query(OntologyTerm).\
filter(OntologyTerm.id.like('GO%')).all()
self.assertEqual(len(terms), len(expected_term_ids))
for i, term in enumerate(terms):
serialized = marshal(term, ontologies_controller.ONT_TERMS_SCHEMA)
self.assertTrue(serialized["id"] in expected_term_ids)
def delete_blocks_test_data():
SESSION.query(SyntenicBlock).delete()
def delete_homologs_test_data():
SESSION.query(Homolog).delete()
def delete_qtls_test_data():
SESSION.query(Feature).delete()
def delete_exons_test_data():
SESSION.query(Exon).delete()
def delete_cytogenetic_band_test_data():
SESSION.query(CytogeneticBand).delete()