def test_format(self):
# This test exists mainly to assert that the directory format is
# defined and functional. More extensive testing is performed
# on its underlying formats (FastqGzFormat).
for read in ['forward', 'reverse']:
filepath = self.get_data_path('%s.fastq.gz' % read)
shutil.copy(filepath,
os.path.join(self.temp_dir.name, '%s.fastq.gz' % read))
format = MultiplexedPairedEndBarcodeInSequenceDirFmt(
self.temp_dir.name, mode='r')
# Should not error.
format.validate()
def setUp(self):
super().setUp()
self.fastq_fp = \
self.get_data_path('forward.fastq.gz')
self.fastq = FastqGzFormat(self.fastq_fp, mode='r')
self.seqs_dir_fmt = MultiplexedPairedEndBarcodeInSequenceDirFmt()
self.seqs_dir_fmt.forward_sequences.write_data(self.fastq,
FastqGzFormat)
self.seqs_dir_fmt.reverse_sequences.write_data(self.fastq,
FastqGzFormat)
self.barcode_series = pd.Series(['A', 'G'],
index=['sample_a', 'sample_b'])
self.per_sample_dir_fmt = SingleLanePerSampleSingleEndFastqDirFmt()
self.untrimmed_dir_fmt = MultiplexedPairedEndBarcodeInSequenceDirFmt()
def demux_paired(seqs: MultiplexedPairedEndBarcodeInSequenceDirFmt,
forward_barcodes: qiime2.CategoricalMetadataColumn,
reverse_barcodes: qiime2.CategoricalMetadataColumn = None,
error_rate: float = 0.1,
batch_size: int = 0,
minimum_length: int = 1,
mixed_orientation: bool = False) -> \
(CasavaOneEightSingleLanePerSampleDirFmt,
MultiplexedPairedEndBarcodeInSequenceDirFmt):
if mixed_orientation and reverse_barcodes is not None:
raise ValueError('Dual-indexed barcodes for mixed orientation '
'reads are not supported.')
per_sample_sequences = CasavaOneEightSingleLanePerSampleDirFmt()
mux_fmt = MultiplexedPairedEndBarcodeInSequenceDirFmt
untrimmed = _demux(
seqs, per_sample_sequences, forward_barcodes, reverse_barcodes,
error_rate, mux_fmt, batch_size, minimum_length)
if mixed_orientation:
fwd = untrimmed.forward_sequences.view(FastqGzFormat)
rev = untrimmed.reverse_sequences.view(FastqGzFormat)
remaining_seqs = MultiplexedPairedEndBarcodeInSequenceDirFmt()
# fwd -> rev && rev -> fwd
remaining_seqs.forward_sequences.write_data(rev, FastqGzFormat)
remaining_seqs.reverse_sequences.write_data(fwd, FastqGzFormat)
untrimmed = _demux(
remaining_seqs, per_sample_sequences, forward_barcodes,
reverse_barcodes, error_rate, mux_fmt, batch_size,
minimum_length)
return per_sample_sequences, untrimmed
def demux_paired(seqs: MultiplexedPairedEndBarcodeInSequenceDirFmt,
forward_barcodes: qiime2.MetadataCategory,
error_rate: float=0.1) -> \
(CasavaOneEightSingleLanePerSampleDirFmt,
MultiplexedPairedEndBarcodeInSequenceDirFmt):
untrimmed = MultiplexedPairedEndBarcodeInSequenceDirFmt()
return _demux(seqs, forward_barcodes, error_rate, untrimmed)