def test_no_labels(self):
"""parse_fasta should return empty list from file w/o seqs"""
# should fail if strict (the default)
self.assertRaises(RecordError, list,
parse_fasta(self.labels, strict=True))
# if not strict, should skip the records
self.assertEqual(list(parse_fasta(self.labels, strict=False)), [])
def test_empty(self):
"""parse_fasta should return empty list from 'file' w/o labels
"""
self.assertEqual(list(parse_fasta(self.empty)), [])
self.assertEqual(list(parse_fasta(self.nolabels, strict=False)),
[])
self.assertRaises(RecordError, list, parse_fasta(self.nolabels))
def test_no_labels(self):
"""parse_fasta should return empty list from file w/o seqs"""
# should fail if strict (the default)
self.assertRaises(RecordError, list,
parse_fasta(self.labels, strict=True))
# if not strict, should skip the records
self.assertEqual(list(parse_fasta(self.labels, strict=False)),
[])
def test_parse_fasta_ignore_comment(self):
"""parse_fasta correct ignores label comments when requested
"""
in_ = '>1\nCAG\n>2 some other info\nCCAG\n>3 \nA'.split('\n')
# ignore_comment = False
actual = list(parse_fasta(in_))
expected = [('1', 'CAG'), ('2 some other info', 'CCAG'), ('3', 'A')]
self.assertEqual(actual, expected)
# ignore_comment = True
actual = list(parse_fasta(in_, ignore_comment=True))
expected = [('1', 'CAG'), ('2', 'CCAG'), ('3', 'A')]
self.assertEqual(actual, expected)
def test_parse_fasta_ignore_comment(self):
"""parse_fasta correct ignores label comments when requested
"""
in_ = '>1\nCAG\n>2 some other info\nCCAG\n>3 \nA'.split('\n')
# ignore_comment = False
actual = list(parse_fasta(in_))
expected = [('1', 'CAG'), ('2 some other info', 'CCAG'), ('3', 'A')]
self.assertEqual(actual, expected)
# ignore_comment = True
actual = list(parse_fasta(in_, ignore_comment=True))
expected = [('1', 'CAG'), ('2', 'CCAG'), ('3', 'A')]
self.assertEqual(actual, expected)
def test_single(self):
"""parse_fasta should read single record as (label, seq) tuple
"""
f = list(parse_fasta(self.oneseq))
self.assertEqual(len(f), 1)
a = f[0]
self.assertEqual(a, ('abc', 'UCAG'))
f = list(parse_fasta(self.multiline))
self.assertEqual(len(f), 1)
a = f[0]
self.assertEqual(a, ('xyz', 'UUUUCCAAAAAG'))
def test_single(self):
"""parse_fasta should read single record as (label, seq) tuple
"""
f = list(parse_fasta(self.oneseq))
self.assertEqual(len(f), 1)
a = f[0]
self.assertEqual(a, ('abc', 'UCAG'))
f = list(parse_fasta(self.multiline))
self.assertEqual(len(f), 1)
a = f[0]
self.assertEqual(a, ('xyz', 'UUUUCCAAAAAG'))
def combine_alignments(fp1, fp2):
"""take two filepointers, combine the files"""
seqs1 = dict(parse_fasta(fp1))
seqs2 = dict(parse_fasta(fp2))
if set(seqs1).intersection(set(seqs2)):
raise ValueError("Conflicting sequence ids in fp1 and fp2")
combined = seqs1
combined.update(seqs2)
return combined
def test_generate_lane_mask(self):
sample_alignment = """>1
AAAAT
>2
AAAGG
>3
AACCC
>4
A----""".split('\n')
aln = Alignment.from_fasta_records(parse_fasta(sample_alignment), DNA)
actual_lanemask = generate_lane_mask(aln, 0.00)
self.assertEqual(actual_lanemask, "11111")
actual_lanemask = generate_lane_mask(aln, 0.10)
self.assertEqual(actual_lanemask, "11100")
actual_lanemask = generate_lane_mask(aln, 0.20)
self.assertEqual(actual_lanemask, "11100")
actual_lanemask = generate_lane_mask(aln, 0.40)
self.assertEqual(actual_lanemask, "11000")
actual_lanemask = generate_lane_mask(aln, 0.60)
self.assertEqual(actual_lanemask, "11000")
actual_lanemask = generate_lane_mask(aln, 0.80)
self.assertEqual(actual_lanemask, "10000")
actual_lanemask = generate_lane_mask(aln, 1.00)
self.assertEqual(actual_lanemask, "00000")
def test_generate_lane_mask(self):
sample_alignment = """>1
AAAAT
>2
AAAGG
>3
AACCC
>4
A----""".split('\n')
aln = Alignment.from_fasta_records(parse_fasta(sample_alignment), DNA)
actual_lanemask = generate_lane_mask(aln, 0.00)
self.assertEqual(actual_lanemask, "11111")
actual_lanemask = generate_lane_mask(aln, 0.10)
self.assertEqual(actual_lanemask, "11100")
actual_lanemask = generate_lane_mask(aln, 0.20)
self.assertEqual(actual_lanemask, "11100")
actual_lanemask = generate_lane_mask(aln, 0.40)
self.assertEqual(actual_lanemask, "11000")
actual_lanemask = generate_lane_mask(aln, 0.60)
self.assertEqual(actual_lanemask, "11000")
actual_lanemask = generate_lane_mask(aln, 0.80)
self.assertEqual(actual_lanemask, "10000")
actual_lanemask = generate_lane_mask(aln, 1.00)
self.assertEqual(actual_lanemask, "00000")
def test_parse_fasta_label_to_name(self):
exp = [('brofist', 'a'), ('brofist', 'caggac'), ('brofist', 'cg')]
# the most powerful fasta label converter known to mankind
obs = list(
parse_fasta(self.threeseq, label_to_name=lambda _: 'brofist'))
self.assertEqual(obs, exp)
def test_multiple(self):
"""parse_fasta should read multiline records correctly"""
f = list(parse_fasta(self.threeseq))
self.assertEqual(len(f), 3)
a, b, c = f
self.assertEqual(a, ('123', 'a'))
self.assertEqual(b, ('abc', 'caggac'))
self.assertEqual(c, ('456', 'cg'))
def test_multiple(self):
"""parse_fasta should read multiline records correctly"""
f = list(parse_fasta(self.threeseq))
self.assertEqual(len(f), 3)
a, b, c = f
self.assertEqual(a, ('123', 'a'))
self.assertEqual(b, ('abc', 'caggac'))
self.assertEqual(c, ('456', 'cg'))
def test_parse_fasta_label_to_name(self):
exp = [('brofist', 'a'), ('brofist', 'caggac'), ('brofist', 'cg')]
# the most powerful fasta label converter known to mankind
obs = list(parse_fasta(self.threeseq,
label_to_name=lambda _: 'brofist'))
self.assertEqual(obs, exp)
def test_gt_bracket_in_seq(self):
"""parse_fasta handles alternate finder function
this test also illustrates how to use the parse_fasta
to handle "sequences" that start with a > symbol, which can
happen when we abuse the parse_fasta to parse
fasta-like sequence quality files.
"""
oneseq_w_gt = '>abc\n>CAG\n'.split('\n')
def get_two_line_records(infile):
line1 = None
for line in infile:
if line1 is None:
line1 = line
else:
yield (line1, line)
line1 = None
f = list(parse_fasta(oneseq_w_gt, finder=get_two_line_records))
self.assertEqual(len(f), 1)
a = f[0]
self.assertEqual(a, ('abc', '>CAG'))
def test_gt_bracket_in_seq(self):
"""parse_fasta handles alternate finder function
this test also illustrates how to use the parse_fasta
to handle "sequences" that start with a > symbol, which can
happen when we abuse the parse_fasta to parse
fasta-like sequence quality files.
"""
oneseq_w_gt = '>abc\n>CAG\n'.split('\n')
def get_two_line_records(infile):
line1 = None
for line in infile:
if line1 is None:
line1 = line
else:
yield (line1, line)
line1 = None
f = list(parse_fasta(oneseq_w_gt, finder=get_two_line_records))
self.assertEqual(len(f), 1)
a = f[0]
self.assertEqual(a, ('abc', '>CAG'))
def test_multiple_bad_strict(self):
with self.assertRaises(RecordError):
list(parse_fasta(self.twogood))
def test_multiple_bad_not_strict(self):
f = list(parse_fasta(self.twogood, strict=False))
self.assertEqual(len(f), 2)
a, b = f
self.assertEqual(a, ('abc', 'caggac'))
def test_multiple_bad_strict(self):
with self.assertRaises(RecordError):
list(parse_fasta(self.twogood))
def test_multiple_bad_not_strict(self):
f = list(parse_fasta(self.twogood, strict=False))
self.assertEqual(len(f), 2)
a, b = f
self.assertEqual(a, ('abc', 'caggac'))
def test_empty(self):
"""parse_fasta should return empty list from 'file' w/o labels
"""
self.assertEqual(list(parse_fasta(self.empty)), [])
self.assertEqual(list(parse_fasta(self.nolabels, strict=False)), [])
self.assertRaises(RecordError, list, parse_fasta(self.nolabels))