def test_join_pairs(self):
with redirected_stdio(stderr=os.devnull):
obs = join_pairs(self.input_seqs)
# manifest is as expected
self._test_manifest(obs)
# expected number of fastq files are created
output_fastqs = list(obs.sequences.iter_views(FastqGzFormat))
self.assertEqual(len(output_fastqs), 3)
# The following values were determined by running vsearch directly
# with default parameters. It is possible that different versions of
# vsearch will result in differences in these numbers, and that
# the corresponding tests may therefore be too specific. We'll have
# to adjust the tests if that's the case.
default_exp_sequence_counts = {
'BAQ2687.1_0_L001_R1_001.fastq.gz': 806,
'BAQ3473.2_1_L001_R1_001.fastq.gz': 753,
'BAQ4697.2_2_L001_R1_001.fastq.gz': 711,
}
for fastq_name, fastq_path in output_fastqs:
seqs = skbio.io.read(str(fastq_path),
format='fastq',
compression='gzip',
constructor=skbio.DNA)
seqs = list(seqs)
seq_lengths = np.asarray([len(s) for s in seqs])
self._test_seq_lengths(seq_lengths)
# expected number of sequences are joined
self.assertEqual(len(seq_lengths),
default_exp_sequence_counts[str(fastq_name)])
def test_join_pairs_some_samples_w_no_joined_seqs(self):
# minmergelen is set very high here, resulting in only one sequence
# being joined across the three samples.
with redirected_stdio(stderr=os.devnull):
obs = join_pairs(self.input_seqs, minmergelen=279)
# manifest is as expected
self._test_manifest(obs)
# expected number of fastq files are created
output_fastqs = list(obs.sequences.iter_views(FastqGzFormat))
self.assertEqual(len(output_fastqs), 3)
# The following values were determined by running vsearch directly.
exp_sequence_counts = {
'BAQ2687.1_0_L001_R1_001.fastq.gz': 0,
'BAQ3473.2_1_L001_R1_001.fastq.gz': 2,
'BAQ4697.2_2_L001_R1_001.fastq.gz': 0,
}
for fastq_name, fastq_path in output_fastqs:
with redirected_stdio(stderr=os.devnull):
seqs = skbio.io.read(str(fastq_path),
format='fastq',
compression='gzip',
constructor=skbio.DNA)
seqs = list(seqs)
seq_lengths = np.asarray([len(s) for s in seqs])
# expected number of sequences are joined
self.assertEqual(len(seq_lengths),
exp_sequence_counts[str(fastq_name)])
def test_join_pairs_all_samples_w_no_joined_seqs(self):
# minmergelen is set very high here, resulting in no sequences
# being joined across the three samples.
with redirected_stdio(stderr=os.devnull):
obs = join_pairs(self.input_seqs, minmergelen=500)
# manifest is as expected
self._test_manifest(obs)
# expected number of fastq files are created
output_fastqs = list(obs.sequences.iter_views(FastqGzFormat))
self.assertEqual(len(output_fastqs), 3)
for fastq_name, fastq_path in output_fastqs:
with redirected_stdio(stderr=os.devnull):
seqs = skbio.io.read(str(fastq_path),
format='fastq',
compression='gzip',
constructor=skbio.DNA)
seqs = list(seqs)
seq_lengths = np.asarray([len(s) for s in seqs])
self.assertEqual(len(seq_lengths), 0)