def populated_db(chanjo_db, exon_lines):
exon_lines = list(exon_lines)
result = link_elements(exon_lines)
chanjo_db.add(*result.models)
results = [
load_transcripts(exon_lines, sample_id="sample", group_id="group"),
load_transcripts(exon_lines, sample_id="sample2", group_id="group"),
]
for result in results:
chanjo_db.add(result.sample)
chanjo_db.add(*result.models)
chanjo_db.save()
yield chanjo_db
def populated_db(chanjo_db, exon_lines):
exon_lines = list(exon_lines)
result = link_elements(exon_lines)
chanjo_db.add(*result.models)
results = [
load_transcripts(exon_lines, sample_id='sample', group_id='group'),
load_transcripts(exon_lines, sample_id='sample2', group_id='group')
]
for result in results:
chanjo_db.add(result.sample)
chanjo_db.add(*result.models)
chanjo_db.save()
yield chanjo_db
def test_load_transcripts(exon_lines):
# GIVEN sambamba depth output lines
# WHEN loading transcript stats
result = sambamba.load_transcripts(exon_lines, sample_id='sample',
group_id='group')
# THEN transcript models should be generated
assert result.count == 10
assert result.sample.id == 'sample'
assert result.sample.group_id == 'group'
assert isinstance(list(result.models)[0], TranscriptStat)
# GIVEN no explicit sample id
# WHEN loading transcript stats
result = sambamba.load_transcripts(exon_lines)
# THEN it should be picked up from file
assert result.sample.id == 'ADM992A10'
def load(context, sample, group, name, group_name, threshold, bed_stream):
"""Load Sambamba output into the database for a sample."""
chanjo_db = ChanjoDB(uri=context.obj['database'])
source = os.path.abspath(bed_stream.name)
result = load_transcripts(bed_stream,
sample_id=sample,
group_id=group,
source=source,
threshold=threshold)
result.sample.name = name
result.sample.group_name = group_name
try:
chanjo_db.add(result.sample)
with click.progressbar(result.models,
length=result.count,
label='loading transcripts') as bar:
for tx_model in bar:
chanjo_db.add(tx_model)
chanjo_db.save()
except IntegrityError as error:
LOG.error('sample already loaded, rolling back')
LOG.debug(error.args[0])
chanjo_db.session.rollback()
context.abort()
def test_load_transcripts(exon_lines):
# GIVEN sambamba depth output lines
# WHEN loading transcript stats
result = sambamba.load_transcripts(exon_lines,
sample_id='sample',
group_id='group')
# THEN transcript models should be generated
assert result.count == 9
assert result.sample.id == 'sample'
assert result.sample.group_id == 'group'
assert isinstance(list(result.models)[0], TranscriptStat)
# GIVEN no explicit sample id
# WHEN loading transcript stats
result = sambamba.load_transcripts(exon_lines)
# THEN it should be picked up from file
assert result.sample.id == 'ADM992A10'
def popexist_db(existing_db, exon_lines):
result = link_elements(exon_lines)
existing_db.add(*result.models)
result = load_transcripts(exon_lines, sample_id="sample", group_id="group")
existing_db.add(result.sample)
existing_db.add(*result.models)
existing_db.save()
yield existing_db
def popexist_db(existing_db, exon_lines):
result = link_elements(exon_lines)
existing_db.add(*result.models)
result = load_transcripts(exon_lines, sample_id='sample', group_id='group')
existing_db.add(result.sample)
existing_db.add(*result.models)
existing_db.save()
yield existing_db
def test_load_transcripts_with_threshold(exon_lines):
# GIVEN a cutoff for "complete" exons at 100x
threshold = 100
# WHEN loading transcript stats
result = sambamba.load_transcripts(exon_lines, threshold=threshold)
# THEN some transcripts will have incomplete exons linked
incompletes = [transcript for transcript in result.models
if transcript.incomplete_exons]
assert len(incompletes) > 0
def test_load_transcripts_with_threshold(exon_lines):
# GIVEN a cutoff for "complete" exons at 100x
threshold = 100
# WHEN loading transcript stats
result = sambamba.load_transcripts(exon_lines, threshold=threshold)
# THEN some transcripts will have incomplete exons linked
incompletes = [
transcript for transcript in result.models
if transcript.incomplete_exons
]
assert len(incompletes) > 0
def upload(self, sample_id: str, sample_name: str, group_id: str, group_name: str,
bed_stream: io.TextIOWrapper):
"""Upload coverage for a sample."""
source = str(Path(bed_stream.name).absolute())
result = load_transcripts(bed_stream, sample_id=sample_id, group_id=group_id,
source=source, threshold=10)
result.sample.name = sample_name
result.sample.group_name = group_name
try:
self.add(result.sample)
with click.progressbar(result.models, length=result.count,
label=f"loading {sample_id}") as progress_bar:
for tx_model in progress_bar:
self.add(tx_model)
self.save()
except IntegrityError as error:
self.session.rollback()
raise error
def load(context, sample, group, name, group_name, threshold, bed_stream):
"""Load Sambamba output into the database for a sample."""
chanjo_db = ChanjoDB(uri=context.obj['database'])
source = os.path.abspath(bed_stream.name)
result = load_transcripts(bed_stream, sample_id=sample, group_id=group,
source=source, threshold=threshold)
result.sample.name = name
result.sample.group_name = group_name
try:
chanjo_db.add(result.sample)
with click.progressbar(result.models, length=result.count,
label='loading transcripts') as bar:
for tx_model in bar:
chanjo_db.add(tx_model)
chanjo_db.save()
except IntegrityError as error:
LOG.error('sample already loaded, rolling back')
LOG.debug(error.args[0])
chanjo_db.session.rollback()
context.abort()