def _run_bax_to_bam(input_file_name, output_file_name):
base_name = ".".join(output_file_name.split(".")[:-2])
input_file_name_tmp = input_file_name
# XXX bax2bam won't write an hdfsubreadset unless the input is XML too
if input_file_name.endswith(".bax.h5"):
input_file_name_tmp = tempfile.NamedTemporaryFile(
suffix=".hdfsubreadset.xml").name
ds_tmp = HdfSubreadSet(input_file_name)
ds_tmp.write(input_file_name_tmp)
args =[
"bax2bam",
"--subread",
"-o", base_name,
"--output-xml", output_file_name,
"--xml", input_file_name_tmp
]
log.info(" ".join(args))
result = run_cmd(" ".join(args),
stdout_fh=sys.stdout,
stderr_fh=sys.stderr)
if result.exit_code != 0:
return result.exit_code
with SubreadSet(output_file_name) as ds:
ds.assertIndexed()
return 0
def _get_bax2bam_inputs():
"""Little hackery to get the setup class Inputs and to avoid calls to
setupclass if skiptest is used
Nat: we want to test that this behaves properly when multiple movies are
supplied as input, so we make an HdfSubreadSet on the fly from various
bax files in testdata
"""
if HAVE_DATA_AND_BAX2BAM:
hdf_subread_xml = tempfile.NamedTemporaryFile(
suffix=".hdfsubreadset.xml").name
bax_files = (
SIV_DATA_DIR +
"/SA3-RS/lambda/2372215/0007_tiny/Analysis_Results/m150404_101626_42267_c100807920800000001823174110291514_s1_p0.1.bax.h5",
pbcore.data.getBaxH5_v23()[0])
ds = HdfSubreadSet(*bax_files)
ds.name = "lambda_rsii"
assert len(set([f.movieName for f in ds.resourceReaders()])) == 2
ds.write(hdf_subread_xml)
return [hdf_subread_xml]
else:
# Assume the test data isn't found and the test won't be run
return ["/path/to/this-test-should-be-skipped.txt"]
def _run_bax_to_bam(input_file_name, output_file_name):
base_name = ".".join(output_file_name.split(".")[:-2])
input_file_name_tmp = input_file_name
# XXX bax2bam won't write an hdfsubreadset unless the input is XML too
if input_file_name.endswith(".bax.h5"):
input_file_name_tmp = tempfile.NamedTemporaryFile(
suffix=".hdfsubreadset.xml").name
ds_tmp = HdfSubreadSet(input_file_name)
ds_tmp.write(input_file_name_tmp)
args =[
"bax2bam",
"--subread",
"-o", base_name,
"--output-xml", output_file_name,
"--xml", input_file_name_tmp
]
log.info(" ".join(args))
result = run_cmd(" ".join(args),
stdout_fh=sys.stdout,
stderr_fh=sys.stderr)
if result.exit_code != 0:
return result.exit_code
with SubreadSet(output_file_name) as ds:
ds.assertIndexed()
return 0
def _run_bax_to_bam(input_file_name, output_file_name):
base_name = ".".join(output_file_name.split(".")[:-2])
input_file_name_tmp = input_file_name
# XXX bax2bam won't write an hdfsubreadset unless the input is XML too
if input_file_name.endswith(".bax.h5"):
input_file_name_tmp = tempfile.NamedTemporaryFile(
suffix=".hdfsubreadset.xml").name
ds_tmp = HdfSubreadSet(input_file_name)
ds_tmp.write(input_file_name_tmp)
args =[
"bax2bam",
"--subread",
"-o", base_name,
"--output-xml", output_file_name,
"--xml", input_file_name_tmp
]
logging.info(" ".join(args))
result = run_cmd(" ".join(args),
stdout_fh=sys.stdout,
stderr_fh=sys.stderr)
if result.exit_code != 0:
return result.exit_code
tmp = tempfile.NamedTemporaryFile(suffix=".subreadset.xml").name
shutil.move(output_file_name, tmp)
# FIXME it would be better to leave this to bax2bam
with SubreadSet(tmp) as ds:
if not ds.isIndexed:
ds.induceIndices()
ds.write(output_file_name)
return 0
def to_chunked_hdfsubreadset_files(hdfsubreadset_path, max_total_nchunks,
chunk_key, dir_name, base_name, ext):
dset = HdfSubreadSet(hdfsubreadset_path, strict=True)
dset_chunks = dset.split(chunks=max_total_nchunks, ignoreSubDatasets=True)
d = {}
for i, dset in enumerate(dset_chunks):
chunk_id = '_'.join([base_name, str(i)])
chunk_name = '.'.join([chunk_id, ext])
chunk_path = os.path.join(dir_name, chunk_name)
dset.write(chunk_path)
d[chunk_key] = os.path.abspath(chunk_path)
c = PipelineChunk(chunk_id, **d)
yield c
def to_chunked_hdfsubreadset_files(hdfsubreadset_path, max_total_nchunks,
chunk_key, dir_name, base_name, ext):
dset = HdfSubreadSet(hdfsubreadset_path, strict=True)
dset_chunks = dset.split(chunks=max_total_nchunks, ignoreSubDatasets=True)
d = {}
for i, dset in enumerate(dset_chunks):
chunk_id = '_'.join([base_name, str(i)])
chunk_name = '.'.join([chunk_id, ext])
chunk_path = os.path.join(dir_name, chunk_name)
dset.write(chunk_path)
d[chunk_key] = os.path.abspath(chunk_path)
c = PipelineChunk(chunk_id, **d)
yield c
def _get_bax2bam_inputs():
"""Little hackery to get the setup class Inputs and to avoid calls to
setupclass if skiptest is used
Nat: we want to test that this behaves properly when multiple movies are
supplied as input, so we make an HdfSubreadSet on the fly from various
bax files in testdata
"""
if HAVE_DATA_AND_BAX2BAM:
hdf_subread_xml = tempfile.NamedTemporaryFile(suffix=".hdfsubreadset.xml").name
bax_files = (SIV_DATA_DIR + "/SA3-RS/lambda/2372215/0007_tiny/Analysis_Results/m150404_101626_42267_c100807920800000001823174110291514_s1_p0.1.bax.h5",
pbtestdata.get_file("rsii-bax-h5"))
ds = HdfSubreadSet(*bax_files)
ds.name = "lambda_rsii"
assert len(set([f.movieName for f in ds.resourceReaders()])) == 2
ds.write(hdf_subread_xml)
return [hdf_subread_xml]
else:
# Assume the test data isn't found and the test won't be run
return ["/path/to/this-test-should-be-skipped.txt"]
