def test_integration(self):
ccs_barcoded = pbtestdata.get_file("ccs-barcoded")
datastore = tempfile.NamedTemporaryFile(suffix=".datastore.json").name
lima_out = tempfile.NamedTemporaryFile(
suffix=".consensusreadset.xml").name
ccs_in = tempfile.NamedTemporaryFile(
suffix=".consensusreadset.xml").name
with ConsensusReadSet(ccs_barcoded) as ccs_tmp:
ccs_tmp.name = "My Data (filtered)"
ccs_tmp.tags = "ccs,filtered"
ccs_tmp.write(ccs_in)
ccs_tmp.name = "lima out"
ccs_tmp.write(lima_out)
ds = DataStore([
DataStoreFile(uuid.uuid4(), "lima", FileTypes.DS_CCS.file_type_id,
lima_out)
])
ds.write_json(datastore)
args = [
"python3", "-m", "pbcoretools.tasks.make_trimmed_dataset",
datastore, ccs_in
]
self._check_call(args)
with ConsensusReadSet("trimmed.consensusreadset.xml",
trustCounts=True) as ccs_out:
assert ccs_out.numRecords > 0
assert ccs_out.name == "My Data (trimmed)"
assert ccs_out.tags == "ccs"
def run_args(args):
dstore = DataStore.load_from_json(os.path.realpath(args.datastore))
ds_in = ConsensusReadSet(args.ccs_in, trustCounts=True)
ds_out = ConsensusReadSet(*([f.path for f in dstore.files.values()]),
trustCounts=True)
sanitize_dataset_tags(ds_out, remove_hidden=True)
ds_out.name = ds_in.name.replace(" (filtered)", "") + " (trimmed)"
ds_out.subdatasets = []
ds_out.write("trimmed.consensusreadset.xml")
return 0
def test_ccs_barcodes_table_asymmetric(self):
CCS_DS = op.join(ROOT_DATA_DIR, "ccs", "asym_barcodes",
"ccs.consensusreadset.xml")
ds = ConsensusReadSet(CCS_DS)
r = to_report(ds, tempfile.mkdtemp())
self.assertEqual(r.tables[1].columns[0].values,
['F5--R5', 'F8--R8', 'F20--R20', 'F29--R29', 'F30--R30'])
def test_ccs_mulitple_movies_single_bam(self):
"""
Check that the report doesn't crash when a single BAM file contains
reads from multiple movies
"""
ds = ConsensusReadSet(self.CCS_BAM)
r = to_report(ds, tempfile.mkdtemp())
def test_ccs_barcodes_table(self):
CCS_DS = pbtestdata.get_file("ccs-barcoded")
ds = ConsensusReadSet(CCS_DS)
r = to_report(ds, tempfile.mkdtemp())
self.assertEqual([c.values for c in r.tables[1].columns[0:4]],
[["lbc1--lbc1", "lbc3--lbc3"], [1, 1], [1958, 1954], [1958, 1954]])
self.assertAlmostEqual(r.tables[1].columns[4].values[0], 0.9724,
places=4)
self.assertAlmostEqual(r.tables[1].columns[4].values[1], 0.9926,
places=4)
def test_get_bio_sample_name(self):
filename = pbtestdata.get_file("subreads-sequel")
ds1 = SubreadSet(filename)
get_bio_sample_name(ds1) == "Narwhale"
filename = pbtestdata.get_file("subreads-biosample-2")
ds2 = SubreadSet(filename)
get_bio_sample_name(ds2) == "UnnamedSample"
ds3 = ds1 + ds2
get_bio_sample_name(ds3) == "Multiple"
filename = pbtestdata.get_file("rsii-ccs-multi-cell")
ds4 = ConsensusReadSet(filename)
get_bio_sample_name(ds4) == "Multiple"
filename = pbtestdata.get_file("ccs-sequel")
ds4 = ConsensusReadSet(filename)
get_bio_sample_name(ds4) == "NarwhalCcs"
def run_dev_ccs_report(rtc):
from pbcore.io import ConsensusReadSet
with ConsensusReadSet(rtc.task.input_files[0]) as ds:
ds.updateCounts()
attr = [
Attribute("number_of_records", value=ds.numRecords),
Attribute("total_length", value=ds.totalLength)
]
report = Report("ccs_report",
title="ConsensusReadSet XML Report",
attributes=attr)
report.write_json(rtc.task.output_files[0])
return 0
def add_bash5(self, filename):
"""Add a bas.h5/ccs.h5/ccs.bam to cacher."""
basename = os.path.basename(filename)
if filename.endswith('.bax.h5'):
movie = basename[:-9]
if movie not in self.bas_files:
self.bas_files[movie] = smrt_wrapper(filename[:-9],
suffix='.bax.h5')
elif filename.endswith('.1.ccs.h5') or \
filename.endswith('.2.ccs.h5') or \
filename.endswith('.3.ccs.h5'):
movie = basename[:-9]
if movie not in self.bas_files:
self.bas_files[movie] = smrt_wrapper(filename[:-9])
elif filename.endswith('.ccs.h5'):
# a single .ccs.h5 (post 150k runs), treat the same as .bas.h5
movie = basename[:-7]
self.bas_files[movie] = defaultdict(lambda: filename)
elif filename.endswith('.1.subreads.bam') or \
filename.endswith('.2.subreads.bam') or \
filename.endswith('.3.subreads.bam'):
movie = basename[:-15]
if movie not in self.bas_files:
self.bas_files[movie] = smrt_wrapper(filename[:-15])
elif filename.endswith('subreads.bam'):
raise NotImplementedError(
"%s add_bash5 *.subreads.bam not implemented." %
(self.__class__.__name__))
elif filename.endswith('.1.ccs.bam') or \
filename.endswith('.2.ccs.bam') or \
filename.endswith('.3.ccs.bam'):
movie = basename[:-10]
if movie not in self.bas_files:
self.bas_files[movie] = smrt_wrapper(filename[:-10])
elif filename.endswith('.bas.h5'):
movie = basename[:-7]
self.bas_files[movie] = defaultdict(lambda: filename)
elif filename.endswith(".consensusreadset.xml"):
ds = ConsensusReadSet(filename)
for rr in ds.resourceReaders():
for rg in rr.readGroupTable:
self.bas_files[rg.MovieName] = dataset_wrapper(filename)
else:
raise IOError("Unsupported file format: %s" % filename)
