def _compute_seqs_per_file(self, input_fasta_fp, num_jobs_to_start):
""" Compute the number of sequences to include in each split file
"""
# count the number of sequences in the fasta file
num_input_seqs = count_seqs(input_fasta_fp)[0]
# divide the number of sequences by the number of jobs to start
result = num_input_seqs / num_jobs_to_start
# if we don't have a perfect split, round up
if result % 1 != 0:
result += 1
# return the result as an integer
return int(result)
def _compute_seqs_per_file(self, input_fasta_fp, num_jobs_to_start):
""" Compute the number of sequences to include in each split file
"""
# count the number of sequences in the fasta file
num_input_seqs = count_seqs(input_fasta_fp)[0]
# divide the number of sequences by the number of jobs to start
result = num_input_seqs / num_jobs_to_start
# if we don't have a perfect split, round up
if result % 1 != 0:
result += 1
# return the result as an integer
return int(result)
def test_run_pick_de_novo_otus_muscle(self):
"""run_pick_de_novo_otus w muscle generates expected results
"""
self.params['assign_taxonomy'] = \
{'id_to_taxonomy_fp': self.test_data['refseqs_tax'][0],
'reference_seqs_fp': self.test_data['refseqs'][0]}
self.params['align_seqs'] = {'alignment_method': 'muscle'}
self.params['filter_alignment'] = \
{'suppress_lane_mask_filter': None,
'entropy_threshold': '0.10'}
run_pick_de_novo_otus(
self.test_data['seqs'][0],
self.test_out,
call_commands_serially,
self.params,
self.qiime_config,
parallel=False,
status_update_callback=no_status_updates)
input_file_basename = splitext(split(self.test_data['seqs'][0])[1])[0]
otu_map_fp = join(self.test_out, 'uclust_picked_otus',
'%s_otus.txt' % input_file_basename)
alignment_fp = join(self.test_out,
'muscle_aligned_seqs', '%s_rep_set_aligned.fasta' %
input_file_basename)
taxonomy_assignments_fp = join(self.test_out,
'uclust_assigned_taxonomy', '%s_rep_set_tax_assignments.txt' %
input_file_basename)
otu_table_fp = join(self.test_out, 'otu_table.biom')
tree_fp = join(self.test_out, 'rep_set.tre')
# Number of OTUs falls within a range that was manually
# confirmed
otu_map_lines = list(open(otu_map_fp))
num_otus = len(otu_map_lines)
otu_map_otu_ids = [o.split()[0] for o in otu_map_lines]
self.assertEqual(num_otus, 14)
# all otus get taxonomy assignments
taxonomy_assignment_lines = list(open(taxonomy_assignments_fp))
self.assertEqual(len(taxonomy_assignment_lines), num_otus)
# all OTUs align
self.assertEqual(count_seqs(alignment_fp)[0], num_otus)
# all OTUs in tree
with open(tree_fp) as f:
tree = TreeNode.from_newick(f)
self.assertEqual(len(list(tree.tips())), num_otus)
# check that the two final output files have non-zero size
self.assertTrue(getsize(tree_fp) > 0)
self.assertTrue(getsize(otu_table_fp) > 0)
# Check that the log file is created and has size > 0
log_fp = glob(join(self.test_out, 'log*.txt'))[0]
self.assertTrue(getsize(log_fp) > 0)
# parse the otu table
otu_table = load_table(otu_table_fp)
expected_sample_ids = [
'f1',
'f2',
'f3',
'f4',
'p1',
'p2',
't1',
't2',
'not16S.1']
# sample IDs are as expected
self.assertItemsEqual(otu_table.ids(), expected_sample_ids)
# expected OTUs
self.assertItemsEqual(otu_table.ids(axis='observation'),
otu_map_otu_ids)
# number of sequences in the full otu table equals the number of
# input sequences
number_seqs_in_otu_table = sum([v.sum()
for v in otu_table.iter_data()])
self.assertEqual(number_seqs_in_otu_table, count_seqs(self.test_data['seqs'][0])[0])
def test_run_pick_de_novo_otus_parallel(self):
"""run_pick_de_novo_otus generates expected results in parallel
"""
self.params['assign_taxonomy'] = \
{'id_to_taxonomy_fp': self.test_data['refseqs_tax'][0],
'reference_seqs_fp': self.test_data['refseqs'][0]}
self.params['align_seqs'] = \
{'template_fp': self.test_data['refseqs_aligned'][0]}
self.params['filter_alignment'] = \
{'lane_mask_fp': self.test_data['refseqs_aligned_lanemask'][0]}
actual_tree_fp, actual_otu_table_fp = run_pick_de_novo_otus(
self.test_data['seqs'][0],
self.test_out,
call_commands_serially,
self.params,
self.qiime_config,
parallel=True,
status_update_callback=no_status_updates)
input_file_basename = splitext(split(self.test_data['seqs'][0])[1])[0]
otu_map_fp = join(self.test_out, 'uclust_picked_otus',
'%s_otus.txt' % input_file_basename)
alignment_fp = join(self.test_out,
'pynast_aligned_seqs', '%s_rep_set_aligned.fasta' %
input_file_basename)
failures_fp = join(self.test_out,
'pynast_aligned_seqs', '%s_rep_set_failures.fasta' %
input_file_basename)
taxonomy_assignments_fp = join(self.test_out,
'uclust_assigned_taxonomy', '%s_rep_set_tax_assignments.txt' %
input_file_basename)
otu_table_fp = join(self.test_out, 'otu_table.biom')
tree_fp = join(self.test_out, 'rep_set.tre')
self.assertEqual(actual_tree_fp, tree_fp)
self.assertEqual(actual_otu_table_fp, otu_table_fp)
# Number of OTUs falls within a range that was manually
# confirmed
otu_map_lines = list(open(otu_map_fp))
num_otus = len(otu_map_lines)
otu_map_otu_ids = [o.split()[0] for o in otu_map_lines]
self.assertEqual(num_otus, 14)
# all otus get taxonomy assignments
taxonomy_assignment_lines = list(open(taxonomy_assignments_fp))
self.assertEqual(len(taxonomy_assignment_lines), num_otus)
# number of seqs which aligned + num of seqs which failed to
# align sum to the number of OTUs
self.assertEqual(count_seqs(alignment_fp)[0] + count_seqs(failures_fp)[0], num_otus)
# number of tips in the tree equals the number of sequences that
# aligned
with open(tree_fp) as f:
tree = TreeNode.from_newick(f)
self.assertEqual(len(list(tree.tips())), count_seqs(alignment_fp)[0])
# parse the otu table
otu_table = load_table(otu_table_fp)
expected_sample_ids = [
'f1',
'f2',
'f3',
'f4',
'p1',
'p2',
't1',
't2',
'not16S.1']
# sample IDs are as expected
self.assertItemsEqual(otu_table.ids(), expected_sample_ids)
# otu ids are as expected
self.assertItemsEqual(otu_table.ids(axis='observation'),
otu_map_otu_ids)
# number of sequences in the full otu table equals the number of
# input sequences
number_seqs_in_otu_table = sum([v.sum()
for v in otu_table.iter_data()])
self.assertEqual(number_seqs_in_otu_table, count_seqs(self.test_data['seqs'][0])[0])
# Check that the log file is created and has size > 0
log_fp = glob(join(self.test_out, 'log*.txt'))[0]
self.assertTrue(getsize(log_fp) > 0)
def test_run_pick_de_novo_otus_muscle(self):
"""run_pick_de_novo_otus w muscle generates expected results
"""
self.params['assign_taxonomy'] = \
{'id_to_taxonomy_fp': self.test_data['refseqs_tax'][0],
'reference_seqs_fp': self.test_data['refseqs'][0]}
self.params['align_seqs'] = {'alignment_method': 'muscle'}
self.params['filter_alignment'] = \
{'suppress_lane_mask_filter': None,
'entropy_threshold': '0.10'}
run_pick_de_novo_otus(self.test_data['seqs'][0],
self.test_out,
call_commands_serially,
self.params,
self.qiime_config,
parallel=False,
status_update_callback=no_status_updates)
input_file_basename = splitext(split(self.test_data['seqs'][0])[1])[0]
otu_map_fp = join(self.test_out, 'uclust_picked_otus',
'%s_otus.txt' % input_file_basename)
alignment_fp = join(self.test_out, 'muscle_aligned_seqs',
'%s_rep_set_aligned.fasta' % input_file_basename)
taxonomy_assignments_fp = join(
self.test_out, 'uclust_assigned_taxonomy',
'%s_rep_set_tax_assignments.txt' % input_file_basename)
otu_table_fp = join(self.test_out, 'otu_table.biom')
tree_fp = join(self.test_out, 'rep_set.tre')
# Number of OTUs falls within a range that was manually
# confirmed
otu_map_lines = list(open(otu_map_fp))
num_otus = len(otu_map_lines)
otu_map_otu_ids = [o.split()[0] for o in otu_map_lines]
self.assertEqual(num_otus, 14)
# all otus get taxonomy assignments
taxonomy_assignment_lines = list(open(taxonomy_assignments_fp))
self.assertEqual(len(taxonomy_assignment_lines), num_otus)
# all OTUs align
self.assertEqual(count_seqs(alignment_fp)[0], num_otus)
# all OTUs in tree
with open(tree_fp) as f:
tree = TreeNode.from_newick(f)
self.assertEqual(len(list(tree.tips())), num_otus)
# check that the two final output files have non-zero size
self.assertTrue(getsize(tree_fp) > 0)
self.assertTrue(getsize(otu_table_fp) > 0)
# Check that the log file is created and has size > 0
log_fp = glob(join(self.test_out, 'log*.txt'))[0]
self.assertTrue(getsize(log_fp) > 0)
# parse the otu table
otu_table = load_table(otu_table_fp)
expected_sample_ids = [
'f1', 'f2', 'f3', 'f4', 'p1', 'p2', 't1', 't2', 'not16S.1'
]
# sample IDs are as expected
self.assertItemsEqual(otu_table.ids(), expected_sample_ids)
# expected OTUs
self.assertItemsEqual(otu_table.ids(axis='observation'),
otu_map_otu_ids)
# number of sequences in the full otu table equals the number of
# input sequences
number_seqs_in_otu_table = sum(
[v.sum() for v in otu_table.iter_data()])
self.assertEqual(number_seqs_in_otu_table,
count_seqs(self.test_data['seqs'][0])[0])
def test_run_pick_de_novo_otus_parallel(self):
"""run_pick_de_novo_otus generates expected results in parallel
"""
self.params['assign_taxonomy'] = \
{'id_to_taxonomy_fp': self.test_data['refseqs_tax'][0],
'reference_seqs_fp': self.test_data['refseqs'][0]}
self.params['align_seqs'] = \
{'template_fp': self.test_data['refseqs_aligned'][0]}
self.params['filter_alignment'] = \
{'lane_mask_fp': self.test_data['refseqs_aligned_lanemask'][0]}
actual_tree_fp, actual_otu_table_fp = run_pick_de_novo_otus(
self.test_data['seqs'][0],
self.test_out,
call_commands_serially,
self.params,
self.qiime_config,
parallel=True,
status_update_callback=no_status_updates)
input_file_basename = splitext(split(self.test_data['seqs'][0])[1])[0]
otu_map_fp = join(self.test_out, 'uclust_picked_otus',
'%s_otus.txt' % input_file_basename)
alignment_fp = join(self.test_out, 'pynast_aligned_seqs',
'%s_rep_set_aligned.fasta' % input_file_basename)
failures_fp = join(self.test_out, 'pynast_aligned_seqs',
'%s_rep_set_failures.fasta' % input_file_basename)
taxonomy_assignments_fp = join(
self.test_out, 'uclust_assigned_taxonomy',
'%s_rep_set_tax_assignments.txt' % input_file_basename)
otu_table_fp = join(self.test_out, 'otu_table.biom')
tree_fp = join(self.test_out, 'rep_set.tre')
self.assertEqual(actual_tree_fp, tree_fp)
self.assertEqual(actual_otu_table_fp, otu_table_fp)
# Number of OTUs falls within a range that was manually
# confirmed
otu_map_lines = list(open(otu_map_fp))
num_otus = len(otu_map_lines)
otu_map_otu_ids = [o.split()[0] for o in otu_map_lines]
self.assertEqual(num_otus, 14)
# all otus get taxonomy assignments
taxonomy_assignment_lines = list(open(taxonomy_assignments_fp))
self.assertEqual(len(taxonomy_assignment_lines), num_otus)
# number of seqs which aligned + num of seqs which failed to
# align sum to the number of OTUs
self.assertEqual(
count_seqs(alignment_fp)[0] + count_seqs(failures_fp)[0], num_otus)
# number of tips in the tree equals the number of sequences that
# aligned
with open(tree_fp) as f:
tree = TreeNode.from_newick(f)
self.assertEqual(len(list(tree.tips())), count_seqs(alignment_fp)[0])
# parse the otu table
otu_table = load_table(otu_table_fp)
expected_sample_ids = [
'f1', 'f2', 'f3', 'f4', 'p1', 'p2', 't1', 't2', 'not16S.1'
]
# sample IDs are as expected
self.assertItemsEqual(otu_table.ids(), expected_sample_ids)
# otu ids are as expected
self.assertItemsEqual(otu_table.ids(axis='observation'),
otu_map_otu_ids)
# number of sequences in the full otu table equals the number of
# input sequences
number_seqs_in_otu_table = sum(
[v.sum() for v in otu_table.iter_data()])
self.assertEqual(number_seqs_in_otu_table,
count_seqs(self.test_data['seqs'][0])[0])
# Check that the log file is created and has size > 0
log_fp = glob(join(self.test_out, 'log*.txt'))[0]
self.assertTrue(getsize(log_fp) > 0)
