def test_tag_sort_bam_dash_t_specified_multiple_times():
args = [
"-i",
data_dir + "unsorted.bam",
"-o",
"test_sorted.bam",
"-t",
consts.CELL_BARCODE_TAG_KEY,
"-t",
consts.GENE_NAME_TAG_KEY,
"-t",
consts.MOLECULE_BARCODE_TAG_KEY,
]
return_call = platform.GenericPlatform.tag_sort_bam(args)
assert return_call == 0
tag_keys = [
consts.CELL_BARCODE_TAG_KEY,
consts.GENE_NAME_TAG_KEY,
consts.MOLECULE_BARCODE_TAG_KEY,
]
with pysam.AlignmentFile("test_sorted.bam", "rb") as f:
segments = f.fetch(until_eof=True)
tag_sortable_record_generator = (
bam.TagSortableRecord.from_aligned_segment(s, tag_keys)
for s in segments)
bam.verify_sort(tag_sortable_record_generator, tag_keys)
for f in glob.glob("test_sorted*"):
os.remove(f)
def test_tag_sort_bam_no_tags():
args = ["-i", data_dir + "unsorted.bam", "-o", "test_sorted.bam"]
return_call = platform.GenericPlatform.tag_sort_bam(args)
assert return_call == 0
tag_keys = []
with pysam.AlignmentFile("test_sorted.bam", "rb") as f:
segments = f.fetch(until_eof=True)
tag_sortable_records = (bam.TagSortableRecord.from_aligned_segment(
s, tag_keys) for s in segments)
bam.verify_sort(tag_sortable_records, tag_keys)
for f in glob.glob("test_sorted*"):
os.remove(f)
def verify_bam_sort(cls, args: Iterable = None) -> int:
"""Command line entrypoint for verifying bam is properly sorted by zero or more tags, followed by queryname.
Parameters
----------
args : Iterable[str], optional
arguments list, for testing (see test/test_entrypoints.py for example). The default
value of None, when passed to `parser.parse_args` causes the parser to
read `sys.argv`
Returns
-------
return_call : 0
return call if the program completes successfully
"""
description = "Verifies whether bam is sorted by the list of zero or more tags, followed by query name"
parser = argparse.ArgumentParser(description=description)
parser.add_argument("-i",
"--input_bam",
required=True,
help="input bamfile")
parser.add_argument(
"-t",
"--tags",
nargs="+",
action="append",
help=
"tag(s) to use to verify sorting, separated by space, e.g. -t CB GE UB",
)
if args is not None:
args = parser.parse_args(args)
else:
args = parser.parse_args()
tags = cls.get_tags(args.tags)
with pysam.AlignmentFile(args.input_bam, "rb") as f:
aligned_segments = f.fetch(until_eof=True)
sortable_records = (bam.TagSortableRecord.from_aligned_segment(
r, tags) for r in aligned_segments)
bam.verify_sort(sortable_records, tags)
print("{0} is correctly sorted by {1} and query name".format(
args.input_bam, tags))
return 0
def test_tag_sort_bam():
args = [
'-i', data_dir + 'unsorted.bam', '-o', 'test_sorted.bam', '-t',
consts.CELL_BARCODE_TAG_KEY, consts.GENE_NAME_TAG_KEY,
consts.MOLECULE_BARCODE_TAG_KEY
]
return_call = platform.GenericPlatform.tag_sort_bam(args)
assert return_call == 0
tag_keys = [
consts.CELL_BARCODE_TAG_KEY, consts.GENE_NAME_TAG_KEY,
consts.MOLECULE_BARCODE_TAG_KEY
]
with pysam.AlignmentFile('test_sorted.bam', 'rb') as f:
segments = f.fetch(until_eof=True)
tag_sortable_records = (bam.TagSortableRecord.from_aligned_segment(
s, tag_keys) for s in segments)
bam.verify_sort(tag_sortable_records, tag_keys)
for f in glob.glob('test_sorted*'):
os.remove(f)