def test_e2e_fasta_beginning_end_sites(self, fasta_file, bed_file,
out_file, sequences, argparser):
from_beginning = 3
from_end = 1
arg_sites = [5, 12]
expected_removals = sorted(
set(TEST_BED_SEQUENCE + [s - 1 for s in arg_sites]))
print(expected_removals)
args = argparser(
"-s %s -o %s --mask %s --mask-from-beginning %s --mask-from-end %s --mask-sites %s"
% (fasta_file, out_file, bed_file, from_beginning, from_end,
" ".join(str(s) for s in arg_sites)))
mask.run(args)
output = SeqIO.parse(out_file, "fasta")
for record in output:
reference = str(sequences[record.id].seq)
masked_seq = str(record.seq)
assert masked_seq[:from_beginning] == "N" * from_beginning
assert masked_seq[-from_end:] == "N" * from_end
for idx, site in enumerate(masked_seq[from_beginning:-from_end],
from_beginning):
if idx in expected_removals:
assert site == "N"
else:
assert site == reference[idx]
def test_run_with_mask_sites(self, vcf_file, out_file, argparser,
mp_context):
args = argparser("--mask-sites 2 8 -s %s -o %s" % (vcf_file, out_file))
def check_mask_sites(mask_sites, *args, **kwargs):
# mask-sites are passed to the CLI as one-indexed
assert mask_sites == [1, 7]
mp_context.setattr(mask, "mask_vcf", check_mask_sites)
mask.run(args)
def test_e2e_fasta_mask_invalid(self, fasta_file, out_file, sequences,
argparser):
args = argparser("-s %s -o %s --mask-invalid" % (fasta_file, out_file))
mask.run(args)
output = SeqIO.parse(out_file, "fasta")
for record in output:
reference = str(sequences[record.id].seq)
for idx, site in enumerate(reference):
assert record.seq[
idx] == site if site in VALID_NUCLEOTIDES else "N"
def test_e2e_vcf_minimal(self, vcf_file, bed_file, argparser):
args = argparser("-s %s --mask %s" % (vcf_file, bed_file))
mask.run(args)
with open(vcf_file) as output:
assert output.readline().startswith("##fileformat") # is a VCF
assert output.readline().startswith(
"#CHROM\tPOS\t") # have a header
for line in output.readlines():
site = int(line.split("\t")[1]) # POS column
site = site - 1 # shift to zero-indexed site
assert site not in TEST_BED_SEQUENCE
def test_e2e_fasta_minimal(self, fasta_file, bed_file, sequences,
argparser):
args = argparser("-s %s --mask %s" % (fasta_file, bed_file))
mask.run(args)
output = SeqIO.parse(fasta_file, "fasta")
for record in output:
reference = sequences[record.id].seq
for idx, site in enumerate(record.seq):
if idx in TEST_BED_SEQUENCE:
assert site == "N"
else:
assert site == reference[idx]
def test_run_with_mask_sites_and_mask_file(self, vcf_file, out_file,
bed_file, argparser,
mp_context):
args = argparser("--mask-sites 20 21 --mask %s -s %s -o %s" %
(bed_file, vcf_file, out_file))
def check_mask_sites(mask_sites, *args, **kwargs):
# mask-sites are passed to the CLI as one-indexed
assert mask_sites == sorted(set(TEST_BED_SEQUENCE + [19, 20]))
mp_context.setattr(mask, "mask_vcf", check_mask_sites)
mask.run(args)
def test_run_fasta_mask_from_beginning_or_end(self, fasta_file, out_file,
argparser, mp_context):
args = argparser(
"-s %s -o %s --mask-from-beginning 2 --mask-from-end 3" %
(fasta_file, out_file))
def check_mask_from(*args, mask_from_beginning=0, mask_from_end=0):
assert mask_from_beginning == 2
assert mask_from_end == 3
mp_context.setattr(mask, "mask_fasta", check_mask_from)
mask.run(args)
def test_run_recognize_vcf(self, bed_file, vcf_file, argparser,
mp_context):
"""Ensure we're handling vcf files correctly"""
args = argparser("--mask=%s -s %s --no-cleanup" % (bed_file, vcf_file))
def fail(*args, **kwargs):
assert False, "Called mask_fasta incorrectly"
mp_context.setattr(mask, "mask_vcf", lambda *a, **k: None)
mp_context.setattr(mask, "mask_fasta", fail)
mp_context.setattr(mask, "copyfile", lambda *args: None)
mask.run(args)
def test_run_normal_case(self, bed_file, vcf_file, out_file, argparser,
mp_context):
def check_args(mask_sites, in_file, _out_file, cleanup):
assert mask_sites == TEST_BED_SEQUENCE, "Wrong mask sites provided"
assert in_file == vcf_file, "Incorrect input file provided"
assert _out_file == out_file, "Incorrect output file provided"
assert cleanup is True, "Cleanup erroneously passed in as False"
mp_context.setattr(mask, "mask_vcf", check_args)
args = argparser("--mask=%s --sequences=%s --output=%s" %
(bed_file, vcf_file, out_file))
mask.run(args)
assert os.path.exists(out_file), "Output file incorrectly deleted"
def test_run_respect_no_cleanup(self, bed_file, vcf_file, argparser,
mp_context):
out_file = os.path.join(os.path.dirname(vcf_file),
"masked_" + os.path.basename(vcf_file))
def make_outfile(mask_sites, in_file, out_file, cleanup=True):
assert cleanup == False
open(out_file, "w").close() # need out_file to exist
mp_context.setattr(mask, "mask_vcf", make_outfile)
args = argparser("--mask=%s -s %s -o %s --no-cleanup" %
(bed_file, vcf_file, out_file))
mask.run(args)
assert os.path.exists(out_file), "Output file incorrectly deleted"
def test_run_handle_missing_outfile(self, bed_file, fasta_file, argparser,
mp_context):
args = argparser("--mask=%s -s %s" % (bed_file, fasta_file))
expected_outfile = os.path.join(
os.path.dirname(fasta_file),
"masked_" + os.path.basename(fasta_file))
def check_outfile(mask_sites, in_file, out_file, **kwargs):
assert out_file == expected_outfile
with open(out_file, "w") as fh:
fh.write("test_string")
mp_context.setattr(mask, "mask_fasta", check_outfile)
mask.run(args)
with open(fasta_file) as fh:
assert fh.read() == "test_string"
def test_e2e_vcf_with_options(self, vcf_file, bed_file, out_file,
argparser):
arg_sites = [5, 12, 14]
expected_removals = sorted(
set(TEST_BED_SEQUENCE + [s - 1 for s in arg_sites]))
args = argparser(
"-s %s -o %s --mask %s --mask-sites %s" %
(vcf_file, out_file, bed_file, " ".join(str(s)
for s in arg_sites)))
mask.run(args)
with open(out_file) as output:
assert output.readline().startswith("##fileformat") # is a VCF
assert output.readline().startswith(
"#CHROM\tPOS\t") # have a header
for line in output.readlines():
site = int(line.split("\t")[1]) # POS column
site = site - 1 #re-zero-index the VCF sites
assert site not in expected_removals
def test_run_handle_empty_sequence_file(self, vcf_file, argparser):
open(vcf_file, "w").close()
args = argparser("-s %s --mask-sites 1" % vcf_file)
with pytest.raises(SystemExit):
mask.run(args)
def test_run_vcf_cannot_mask_beginning_or_end(self, vcf_file, argparser,
op):
args = argparser("-s %s --mask-from-%s 2" % (vcf_file, op))
with pytest.raises(SystemExit) as err:
mask.run(args)
def test_run_requires_some_masking(self, vcf_file, argparser):
args = argparser("-s %s" % vcf_file)
with pytest.raises(SystemExit) as err:
mask.run(args)
def test_run_handle_missing_sequence_file(self, vcf_file, argparser):
os.remove(vcf_file)
args = argparser("-s %s" % vcf_file)
with pytest.raises(SystemExit):
mask.run(args)
def test_run_handle_empty_mask_file(self, vcf_file, bed_file, argparser):
open(bed_file, "w").close()
args = argparser("-s %s --mask %s" % (vcf_file, bed_file))
with pytest.raises(SystemExit):
mask.run(args)
def test_run_handle_missing_mask_file(self, vcf_file, bed_file, argparser):
os.remove(bed_file)
args = argparser("-s %s --mask %s" % (vcf_file, bed_file))
with pytest.raises(SystemExit):
mask.run(args)