def test_sanger_variant(self):
# test entire range of possible ascii chars for sanger
all_sanger_ascii = ('!"#$%&\'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOP'
'QRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~')
obs = _decode_qual_to_phred(all_sanger_ascii, variant='sanger')
npt.assert_equal(obs, np.arange(94))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('a b', variant='sanger')
self.assertIn('[0, 93]', str(cm.exception))
def test_sanger_variant(self):
# test entire range of possible ascii chars for sanger
all_sanger_ascii = ('!"#$%&\'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOP'
'QRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~')
obs = _decode_qual_to_phred(all_sanger_ascii, variant='sanger')
npt.assert_equal(obs, np.arange(94))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('a b', variant='sanger')
self.assertIn('[0, 93]', str(cm.exception))
def test_illumina18_variant(self):
# test entire range of possible ascii chars for illumina1.8
all_illumina18_ascii = ('!"#$%&\'()*+,-./0123456789:;<=>?@ABCDEFGHIJKL'
'MNOPQRSTUVWXYZ[\\]^_')
obs = _decode_qual_to_phred(all_illumina18_ascii,
variant='illumina1.8')
npt.assert_equal(obs, np.arange(63))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('AaB', variant='illumina1.8')
self.assertIn('[0, 62]', str(cm.exception))
def test_illumina13_variant(self):
# test entire range of possible ascii chars for illumina1.3
all_illumina13_ascii = ('@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijk'
'lmnopqrstuvwxyz{|}~')
obs = _decode_qual_to_phred(all_illumina13_ascii,
variant='illumina1.3')
npt.assert_equal(obs, np.arange(63))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('a!b', variant='illumina1.3')
self.assertIn('[0, 62]', str(cm.exception))
def test_illumina13_variant(self):
# test entire range of possible ascii chars for illumina1.3
all_illumina13_ascii = ('@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijk'
'lmnopqrstuvwxyz{|}~')
obs = _decode_qual_to_phred(all_illumina13_ascii,
variant='illumina1.3')
npt.assert_equal(obs, np.arange(63))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('a!b', variant='illumina1.3')
self.assertIn('[0, 62]', str(cm.exception))
def test_illumina18_variant(self):
# test entire range of possible ascii chars for illumina1.8
all_illumina18_ascii = ('!"#$%&\'()*+,-./0123456789:;<=>?@ABCDEFGHIJKL'
'MNOPQRSTUVWXYZ[\\]^_')
obs = _decode_qual_to_phred(all_illumina18_ascii,
variant='illumina1.8')
npt.assert_equal(obs, np.arange(63))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('AaB', variant='illumina1.8')
self.assertIn('[0, 62]', str(cm.exception))
def _parse_quality_scores(fh, seq_len, variant, phred_offset, prev):
phred_scores = []
qual_len = 0
for chunk in _line_generator(fh, skip_blanks=False):
if chunk:
if chunk.startswith('@') and qual_len == seq_len:
return np.hstack(phred_scores), chunk
else:
if not prev:
_blank_error("after '+' or within quality scores")
qual_len += len(chunk)
if qual_len > seq_len:
raise FASTQFormatError(
"Found more quality score characters than sequence "
"characters. Extra quality score characters: %r" %
chunk[-(qual_len - seq_len):])
phred_scores.append(
_decode_qual_to_phred(chunk, variant=variant,
phred_offset=phred_offset))
prev = chunk
if qual_len != seq_len:
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
return np.hstack(phred_scores), None
def _qseq_to_generator(fh,
constructor=Sequence,
filter=_will_filter,
phred_offset=_default_phred_offset,
variant=_default_variant,
**kwargs):
for line in fh:
(machine_name, run, lane, tile, x, y, index, read, seq, raw_qual,
filtered) = _record_parser(line)
if not filter or not filtered:
phred = _decode_qual_to_phred(raw_qual, variant, phred_offset)
seq_id = '%s_%s:%s:%s:%s:%s#%s/%s' % (machine_name, run, lane,
tile, x, y, index, read)
yield constructor(seq,
metadata={
'id': seq_id,
'machine_name': machine_name,
'run_number': int(run),
'lane_number': int(lane),
'tile_number': int(tile),
'x': int(x),
'y': int(y),
'index': int(index),
'read_number': int(read)
},
positional_metadata={'quality': phred},
**kwargs)
def _parse_quality_scores(fh, seq_len, variant, phred_offset, prev):
phred_scores = []
qual_len = 0
for chunk in _line_generator(fh, skip_blanks=False):
if chunk:
if chunk.startswith('@') and qual_len == seq_len:
return np.hstack(phred_scores), chunk
else:
if not prev:
_blank_error("after '+' or within quality scores")
qual_len += len(chunk)
if qual_len > seq_len:
raise FASTQFormatError(
"Found more quality score characters than sequence "
"characters. Extra quality score characters: %r" %
chunk[-(qual_len - seq_len):])
phred_scores.append(
_decode_qual_to_phred(chunk, variant=variant,
phred_offset=phred_offset))
prev = chunk
if qual_len != seq_len:
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
return np.hstack(phred_scores), None
def _qseq_to_generator(fh, constructor=Sequence, filter=_will_filter,
phred_offset=_default_phred_offset,
variant=_default_variant):
for line in fh:
(machine_name, run, lane, tile, x, y, index, read, seq, raw_qual,
filtered) = _record_parser(line)
if not filter or not filtered:
phred = _decode_qual_to_phred(raw_qual, variant, phred_offset)
seq_id = '%s_%s:%s:%s:%s:%s#%s/%s' % (
machine_name, run, lane, tile, x, y, index, read)
yield constructor(seq, quality=phred, id=seq_id)
def _qseq_to_generator(fh, constructor=BiologicalSequence, filter=_will_filter,
phred_offset=_default_phred_offset,
variant=_default_variant):
for line in fh:
(machine_name, run, lane, tile, x, y, index, read, seq, raw_qual,
filtered) = _record_parser(line)
if not filter or not filtered:
phred = _decode_qual_to_phred(raw_qual, variant, phred_offset)
seq_id = '%s_%s:%s:%s:%s:%s#%s/%s' % (
machine_name, run, lane, tile, x, y, index, read)
yield constructor(seq, quality=phred, id=seq_id)
def test_custom_phred_offset(self):
ascii_chars = '*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\'
obs = _decode_qual_to_phred(ascii_chars, phred_offset=42)
npt.assert_equal(obs, np.arange(51))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred(ascii_chars, phred_offset=43)
self.assertIn('[0, 83]', str(cm.exception))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred(ascii_chars, phred_offset=0)
self.assertIn('`phred_offset`', str(cm.exception))
self.assertIn('printable', str(cm.exception))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred(ascii_chars, phred_offset=127)
self.assertIn('`phred_offset`', str(cm.exception))
self.assertIn('printable', str(cm.exception))
def test_custom_phred_offset(self):
ascii_chars = '*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\'
obs = _decode_qual_to_phred(ascii_chars, phred_offset=42)
npt.assert_equal(obs, np.arange(51))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred(ascii_chars, phred_offset=43)
self.assertIn('[0, 83]', str(cm.exception))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred(ascii_chars, phred_offset=0)
self.assertIn('`phred_offset`', str(cm.exception))
self.assertIn('printable', str(cm.exception))
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred(ascii_chars, phred_offset=127)
self.assertIn('`phred_offset`', str(cm.exception))
self.assertIn('printable', str(cm.exception))
def _parse_quality_scores(fh, seq_len, variant, phred_offset):
phred_scores = []
qual_len = 0
for chunk in _line_generator(fh):
if chunk.startswith('@') and qual_len == seq_len:
return phred_scores, chunk
else:
qual_len += len(chunk)
if qual_len > seq_len:
raise FASTQFormatError(
"Found more quality score characters than sequence "
"characters. Extra quality score characters: %r" %
chunk[-(qual_len - seq_len):])
phred_scores.extend(
_decode_qual_to_phred(chunk, variant=variant,
phred_offset=phred_offset))
if qual_len != seq_len:
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
return phred_scores, None
def _parse_quality_scores(fh, seq_len, variant, phred_offset):
phred_scores = []
qual_len = 0
for chunk in _line_generator(fh):
if chunk.startswith('@') and qual_len == seq_len:
return phred_scores, chunk
else:
qual_len += len(chunk)
if qual_len > seq_len:
raise FASTQFormatError(
"Found more quality score characters than sequence "
"characters. Extra quality score characters: %r" %
chunk[-(qual_len - seq_len):])
phred_scores.extend(
_decode_qual_to_phred(chunk,
variant=variant,
phred_offset=phred_offset))
if qual_len != seq_len:
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
return phred_scores, None
def test_solexa_variant(self):
with self.assertRaises(NotImplementedError) as cm:
_decode_qual_to_phred('abcd', variant='solexa')
self.assertIn('719', str(cm.exception))
def test_empty_qual_str(self):
self.assertEqual(_decode_qual_to_phred('', variant='sanger'), [])
def test_empty_qual_str(self):
self.assertEqual(_decode_qual_to_phred('', variant='sanger'), [])
def test_empty_qual_str(self):
npt.assert_equal(_decode_qual_to_phred('', variant='sanger'),
np.array([], dtype=np.uint8))
def test_unrecognized_variant(self):
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('abcd', variant='illumina')
self.assertIn('variant', str(cm.exception))
self.assertIn("'illumina'", str(cm.exception))
def test_solexa_variant(self):
with self.assertRaises(NotImplementedError) as cm:
_decode_qual_to_phred('abcd', variant='solexa')
self.assertIn('719', str(cm.exception))
def test_variant_and_phred_offset_provided(self):
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('abcd', variant='sanger', phred_offset=64)
self.assertIn('both', str(cm.exception))
self.assertIn('`variant`', str(cm.exception))
self.assertIn('`phred_offset`', str(cm.exception))
def test_missing_variant_and_phred_offset(self):
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('abcd')
self.assertIn('`variant`', str(cm.exception))
self.assertIn('`phred_offset`', str(cm.exception))
self.assertIn('decode', str(cm.exception))
def test_missing_variant_and_phred_offset(self):
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('abcd')
self.assertIn('`variant`', str(cm.exception))
self.assertIn('`phred_offset`', str(cm.exception))
self.assertIn('decode', str(cm.exception))
def test_empty_qual_str(self):
npt.assert_equal(_decode_qual_to_phred('', variant='sanger'),
np.array([], dtype=np.uint8))
def test_unrecognized_variant(self):
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('abcd', variant='illumina')
self.assertIn('variant', str(cm.exception))
self.assertIn("'illumina'", str(cm.exception))
def test_variant_and_phred_offset_provided(self):
with self.assertRaises(ValueError) as cm:
_decode_qual_to_phred('abcd', variant='sanger', phred_offset=64)
self.assertIn('both', str(cm.exception))
self.assertIn('`variant`', str(cm.exception))
self.assertIn('`phred_offset`', str(cm.exception))
