def stipulate(args):
"""
REQUIRED TO CREATE ITERABLE FUNCTIONS TO RUN IN CUTADAPT 2.7. THIS FUNCTION IS CALLED ONLY ONE TIME.
"""
modifiers = []
pipeline_add = modifiers.append
adapter_parser = AdapterParser(
max_error_rate=args.error_rate,
min_overlap=args.overlap,
read_wildcards=args.match_read_wildcards,
adapter_wildcards=args.match_adapter_wildcards,
indels=args.indels,
)
adapters = adapter_parser.parse_multi(args.adapters)
warn_duplicate_adapters(adapters)
if args.nextseq_trim is not None:
pipeline_add(NextseqQualityTrimmer(args.nextseq_trim, args.phred64))
if args.quality_cutoff is not None:
cutoffs = parse_cutoffs(args.quality_cutoff)
pipeline_add(QualityTrimmer(cutoffs[0], cutoffs[1], args.phred64))
adapter_cutter = None
if adapters:
adapter_cutter = AdapterCutter(adapters, args.times, args.action)
pipeline_add(adapter_cutter)
if args.trim_n:
pipeline_add(NEndTrimmer())
add_unconditional_cutters(pipeline_add, args.cut)
print("modifiers (cutadapt):", modifiers)
return modifiers
def test_parse_file_notation(tmpdir):
tmp_path = str(tmpdir.join('adapters.fasta'))
with open(tmp_path, 'w') as f:
f.write(
dedent(""">first_name
ADAPTER1
>second_name
ADAPTER2
"""))
parser = AdapterParser(max_error_rate=0.2,
min_overlap=4,
read_wildcards=False,
adapter_wildcards=False,
indels=False)
adapters = list(parser.parse('file:' + tmp_path, cmdline_type='back'))
assert len(adapters) == 2
assert adapters[0].name == 'first_name'
assert adapters[0].sequence == 'ADAPTER1'
assert adapters[1].name == 'second_name'
assert adapters[1].sequence == 'ADAPTER2'
for a in adapters:
assert a.max_error_rate == 0.2
assert a.min_overlap == 4
assert not a.read_wildcards
assert not a.adapter_wildcards
assert not a.indels
def test_parse_with_adapter_sequence_as_a_path(tmp_path):
parser = AdapterParser()
with pytest.raises(InvalidCharacter):
parser._parse("invalid.character", "back")
# user forgot to write "file:"
path = (tmp_path / "afile.fasta")
path.write_text(">abc\nACGT\n")
with pytest.raises(InvalidCharacter) as e:
list(parser.parse(str(path), "back"))
assert "A file exists named" in e.value.args[0]
def test_anywhere_parameter_front():
parser = AdapterParser(max_error_rate=0.2, min_overlap=4, read_wildcards=False,
adapter_wildcards=False, indels=True)
adapter = list(parser.parse('CTGAAGTGAAGTACACGGTT;anywhere', 'front'))[0]
assert isinstance(adapter, FrontAdapter)
assert adapter._force_anywhere
# TODO move the rest to a separate test
read = Sequence('foo1', 'AAAAAAAAAACTGAAGTGAA')
from cutadapt.modifiers import AdapterCutter
cutter = AdapterCutter([adapter])
trimmed_read = cutter(read, ModificationInfo(read))
assert trimmed_read.sequence == ''
def test_anywhere_parameter():
parser = AdapterParser(max_error_rate=0.2,
min_overlap=4,
read_wildcards=False,
adapter_wildcards=False,
indels=True)
adapter = list(parser.parse('CTGAAGTGAAGTACACGGTT;anywhere', 'back'))[0]
assert adapter.remove == 'suffix'
assert adapter.where is Where.ANYWHERE
read = Sequence('foo1', 'TGAAGTACACGGTTAAAAAAAAAA')
from cutadapt.modifiers import AdapterCutter
cutter = AdapterCutter([adapter])
trimmed_read = cutter(read, [])
assert trimmed_read.sequence == ''
def test_parse_with_parameters():
parser = AdapterParser(
max_error_rate=0.2, min_overlap=4, read_wildcards=False,
adapter_wildcards=False, indels=False)
a = parser._parse('ACGTACGT; e=0.15', 'front')
assert a.max_error_rate == 0.15
assert a.min_overlap == 4
a = parser._parse('ACGTAAAA; o=5; e=0.11', 'back')
assert a.max_error_rate == 0.11
assert a.min_overlap == 5
for spec in ('thename=ACG;e=0.15 ... TGT;e=0.17', 'thename=ACG;e=0.15...TGT;e=0.17'):
a = parser._parse(spec, 'back')
assert isinstance(a, LinkedAdapter)
assert a.front_adapter.max_error_rate == 0.15
assert a.back_adapter.max_error_rate == 0.17
def test_linked_adapter_parameters():
# issue #394
a = AdapterParser(max_error_rate=0.17, indels=False)._parse("ACG...TGT")
assert isinstance(a, LinkedAdapter)
assert a.front_adapter.max_error_rate == 0.17
assert a.back_adapter.max_error_rate == 0.17
assert not a.front_adapter.indels
assert not a.back_adapter.indels
def test_parse_with_parameters(tmp_path):
parser = AdapterParser(max_errors=0.2,
min_overlap=4,
read_wildcards=False,
adapter_wildcards=False,
indels=False)
a = parser._parse('ACGTACGT; e=0.15', 'front')
assert isinstance(a, FrontAdapter)
assert a.max_error_rate == 0.15
assert a.min_overlap == 4
a = parser._parse('ACGTAAAA; o=5; e=0.11', 'back')
assert isinstance(a, BackAdapter)
assert a.max_error_rate == 0.11
assert a.min_overlap == 5
for spec in ('thename=ACG;e=0.15 ... TGT;e=0.17',
'thename=ACG;e=0.15...TGT;e=0.17'):
a = parser._parse(spec, 'back')
assert isinstance(a, LinkedAdapter)
assert a.front_adapter.max_error_rate == 0.15
assert a.back_adapter.max_error_rate == 0.17
with pytest.raises(ValueError) as e:
parser._parse("A", "invalid-cmdline-type")
assert "cmdline_type cannot be" in e.value.args[0]
def adapters_from_args(args) -> Tuple[List[Adapter], List[Adapter]]:
adapter_parser = AdapterParser(
max_errors=args.error_rate,
min_overlap=args.overlap,
read_wildcards=args.match_read_wildcards,
adapter_wildcards=args.match_adapter_wildcards,
indels=args.indels,
)
try:
adapters = adapter_parser.parse_multi(args.adapters)
adapters2 = adapter_parser.parse_multi(args.adapters2)
except (FileNotFoundError, ValueError, InvalidCharacter) as e:
raise CommandLineError(e)
warn_duplicate_adapters(adapters)
warn_duplicate_adapters(adapters2)
if args.debug == "trace":
for adapter in adapters + adapters2:
adapter.enable_debug()
return adapters, adapters2
def pipeline_from_parsed_args(args, paired, is_interleaved_output):
"""
Setup a processing pipeline from parsed command-line arguments.
If there are any problems parsing the arguments, a CommandLineError is thrown.
Return an instance of Pipeline (SingleEndPipeline or PairedEndPipeline)
"""
check_arguments(args, paired, is_interleaved_output)
if args.action == 'none':
args.action = None
adapter_parser = AdapterParser(
max_error_rate=args.error_rate,
min_overlap=args.overlap,
read_wildcards=args.match_read_wildcards,
adapter_wildcards=args.match_adapter_wildcards,
indels=args.indels,
)
try:
adapters = adapter_parser.parse_multi(args.adapters, args.anywhere, args.front)
adapters2 = adapter_parser.parse_multi(args.adapters2, args.anywhere2, args.front2)
except IOError as e:
if e.errno == errno.ENOENT:
raise CommandLineError(e)
raise
except ValueError as e:
raise CommandLineError(e)
if args.debug:
for adapter in adapters + adapters2:
adapter.enable_debug()
# Create the processing pipeline
if paired:
pair_filter_mode = 'any' if args.pair_filter is None else args.pair_filter
pipeline = PairedEndPipeline(pair_filter_mode)
else:
pipeline = SingleEndPipeline()
# When adapters are being trimmed only in R1 or R2, override the pair filter mode
# as using the default of 'any' would regard all read pairs as untrimmed.
if paired and (not adapters2 or not adapters) and (
args.discard_untrimmed or args.untrimmed_output or args.untrimmed_paired_output):
pipeline.override_untrimmed_pair_filter = True
for i, cut_arg in enumerate([args.cut, args.cut2]):
# cut_arg is a list
if not cut_arg:
continue
if len(cut_arg) > 2:
raise CommandLineError("You cannot remove bases from more than two ends.")
if len(cut_arg) == 2 and cut_arg[0] * cut_arg[1] > 0:
raise CommandLineError("You cannot remove bases from the same end twice.")
for c in cut_arg:
if c == 0:
continue
if i == 0: # R1
if paired:
pipeline.add(UnconditionalCutter(c), None)
else:
pipeline.add(UnconditionalCutter(c))
else:
# R2
assert isinstance(pipeline, PairedEndPipeline)
pipeline.add(None, UnconditionalCutter(c))
pipeline_add = pipeline.add_both if paired else pipeline.add
if args.nextseq_trim is not None:
pipeline_add(NextseqQualityTrimmer(args.nextseq_trim, args.quality_base))
if args.quality_cutoff is not None:
cutoffs = parse_cutoffs(args.quality_cutoff)
pipeline_add(QualityTrimmer(cutoffs[0], cutoffs[1], args.quality_base))
if args.pair_adapters:
if not paired:
raise CommandLineError("Option --pair-adapters can only be used when trimming "
"paired-end reads")
if args.times != 1:
raise CommandLineError("--pair-adapters cannot be used with --times")
try:
cutter = PairedAdapterCutter(adapters, adapters2, args.action)
except PairedAdapterCutterError as e:
raise CommandLineError("--pair-adapters: " + str(e))
pipeline.add_paired_modifier(cutter)
else:
adapter_cutter, adapter_cutter2 = None, None
if adapters:
adapter_cutter = AdapterCutter(adapters, args.times, args.action)
if adapters2:
adapter_cutter2 = AdapterCutter(adapters2, args.times, args.action)
if paired:
if adapter_cutter or adapter_cutter2:
pipeline.add(adapter_cutter, adapter_cutter2)
else:
if adapter_cutter:
pipeline.add(adapter_cutter)
# Remaining modifiers that apply to both reads of paired-end reads
if args.length is not None:
pipeline_add(Shortener(args.length))
if args.trim_n:
pipeline_add(NEndTrimmer())
if args.length_tag:
pipeline_add(LengthTagModifier(args.length_tag))
for suffix in args.strip_suffix:
pipeline_add(SuffixRemover(suffix))
if args.prefix or args.suffix:
pipeline_add(PrefixSuffixAdder(args.prefix, args.suffix))
if args.zero_cap:
pipeline_add(ZeroCapper(quality_base=args.quality_base))
# Set filtering parameters
# Minimum/maximum length
for attr in 'minimum_length', 'maximum_length':
param = getattr(args, attr)
if param is not None:
lengths = parse_lengths(param)
if not paired and len(lengths) == 2:
raise CommandLineError('Two minimum or maximum lengths given for single-end data')
if paired and len(lengths) == 1:
lengths = (lengths[0], lengths[0])
setattr(pipeline, attr, lengths)
pipeline.max_n = args.max_n
pipeline.discard_casava = args.discard_casava
pipeline.discard_trimmed = args.discard_trimmed
pipeline.discard_untrimmed = args.discard_untrimmed
return pipeline
def test_parse_invalid_adapter_specific_parameter(where, reqopt):
parser = AdapterParser()
with pytest.raises(ValueError) as e:
parser._parse_not_linked("A;{}".format(reqopt), "name", where)
assert "can only be used within linked adapters" in e.value.args[0]
def test_linked_adapter_name():
# issue #414
a = AdapterParser()._parse("the_name=^ACG...TGT")
assert isinstance(a, LinkedAdapter)
assert a.create_statistics().name == "the_name"
def test_linked_adapter_front_required_optional(r1, r2, exp1, exp2):
# -g X...Y
a = AdapterParser()._parse("ACG" + r1 + "...TGT" + r2, "front")
assert isinstance(a, LinkedAdapter)
assert a.front_required is exp1
assert a.back_required is exp2
def test_linked_adapter_back_required_optional(r1, r2, req1, req2):
# -a X...Y
a = AdapterParser()._parse("ACG" + r1 + "...TGT" + r2, "back")
assert isinstance(a, LinkedAdapter)
assert a.front_required is req1
assert a.back_required is req2
def test_anchoring_makes_front_linked_adapter_required(seq, req1, req2):
# -a X...Y
a = AdapterParser()._parse(seq, "back")
assert isinstance(a, LinkedAdapter)
assert a.front_required is req1
assert a.back_required is req2
def test_parse_multi():
parser = AdapterParser()
with pytest.raises(ValueError) as e:
parser.parse_multi([("invalid-type", "A")])
assert "adapter type must be" in e.value.args[0]
def pipeline_from_parsed_args(args, paired, is_interleaved_output):
"""
Setup a processing pipeline from parsed command-line arguments.
If there are any problems parsing the arguments, a CommandLineError is raised.
Return an instance of Pipeline (SingleEndPipeline or PairedEndPipeline)
"""
check_arguments(args, paired, is_interleaved_output)
if args.action == 'none':
args.action = None
adapter_parser = AdapterParser(
max_error_rate=args.error_rate,
min_overlap=args.overlap,
read_wildcards=args.match_read_wildcards,
adapter_wildcards=args.match_adapter_wildcards,
indels=args.indels,
)
try:
adapters = adapter_parser.parse_multi(args.adapters)
adapters2 = adapter_parser.parse_multi(args.adapters2)
except (FileNotFoundError, ValueError) as e:
raise CommandLineError(e)
warn_duplicate_adapters(adapters)
warn_duplicate_adapters(adapters2)
if args.debug:
for adapter in adapters + adapters2:
adapter.enable_debug()
# Create the processing pipeline
if paired:
pair_filter_mode = 'any' if args.pair_filter is None else args.pair_filter
pipeline = PairedEndPipeline(pair_filter_mode)
else:
pipeline = SingleEndPipeline()
# When adapters are being trimmed only in R1 or R2, override the pair filter mode
# as using the default of 'any' would regard all read pairs as untrimmed.
if paired and (not adapters2
or not adapters) and (args.discard_untrimmed
or args.untrimmed_output
or args.untrimmed_paired_output):
pipeline.override_untrimmed_pair_filter = True
add_unconditional_cutters(pipeline, args.cut, args.cut2, paired)
pipeline_add = pipeline.add_both if paired else pipeline.add
if args.nextseq_trim is not None:
pipeline_add(
NextseqQualityTrimmer(args.nextseq_trim, args.quality_base))
if args.quality_cutoff is not None:
cutoffs = parse_cutoffs(args.quality_cutoff)
pipeline_add(QualityTrimmer(cutoffs[0], cutoffs[1], args.quality_base))
if args.pair_adapters:
try:
cutter = PairedAdapterCutter(adapters, adapters2, args.action)
except PairedAdapterCutterError as e:
raise CommandLineError("--pair-adapters: " + str(e))
pipeline.add_paired_modifier(cutter)
else:
adapter_cutter, adapter_cutter2 = None, None
if adapters:
adapter_cutter = AdapterCutter(adapters, args.times, args.action)
if adapters2:
adapter_cutter2 = AdapterCutter(adapters2, args.times, args.action)
if paired:
if adapter_cutter or adapter_cutter2:
pipeline.add(adapter_cutter, adapter_cutter2)
else:
if adapter_cutter:
pipeline.add(adapter_cutter)
for modifier in modifiers_applying_to_both_ends_if_paired(args):
pipeline_add(modifier)
# Set filtering parameters
# Minimum/maximum length
for attr in 'minimum_length', 'maximum_length':
param = getattr(args, attr)
if param is not None:
lengths = parse_lengths(param)
if not paired and len(lengths) == 2:
raise CommandLineError(
'Two minimum or maximum lengths given for single-end data')
if paired and len(lengths) == 1:
lengths = (lengths[0], lengths[0])
setattr(pipeline, attr, lengths)
pipeline.max_n = args.max_n
pipeline.discard_casava = args.discard_casava
pipeline.discard_trimmed = args.discard_trimmed
pipeline.discard_untrimmed = args.discard_untrimmed
return pipeline
