def generate_biom_table(seqs_fp,
uc_fp,
delim='_'):
"""Generate BIOM table and representative FASTA set
Parameters
----------
seqs_fp: string
file path to deblurred sequences
uc_fp: string
file path to dereplicated sequences map (.uc format)
delim: string, optional
delimiter for splitting sample and sequence IDs in sequence label
default: '_'
Returns
-------
deblur_clusters: dictionary
dictionary of clusters including dereplicated sequence labels
Table: biom.table
an instance of a BIOM table
"""
# parse clusters in dereplicated sequences map (.uc format)
with open(uc_fp, 'U') as uc_f:
derep_clusters, failures, seeds = clusters_from_uc_file(uc_f)
# parse clusters in deblur file, set observation ID to be the sequence
deblur_clusters = parse_deblur_output(seqs_fp, derep_clusters)
# create sparse dictionary of observation and sample ID counts
data, otu_ids, sample_ids = generate_biom_data(deblur_clusters, delim)
# build BIOM table
return deblur_clusters, Table(data, otu_ids, sample_ids,
observation_metadata=None,
sample_metadata=None, table_id=None,
generated_by="deblur",
create_date=datetime.now().isoformat())
def test_clusters_from_uc_file(self):
""" clusters_from_uc_file functions as expected """
expected_clusters = {'s2': ['s2', 's3']}
expected_failures = ['s1']
expected_new_seeds = ['s2']
self.assertEqual(clusters_from_uc_file(self.uc_lines1),
(expected_clusters, expected_failures, expected_new_seeds))
def test_clusters_from_uc_file_multiple_hits(self):
""" clusters_from_uc_file handles error_on_multiple_hits correctly
"""
# when a query hits multiple hits and error_on_multiple_hits=True
# an error should be raised
self.assertRaises(UclustParseError,
clusters_from_uc_file,
self.uc_lines_w_multiple_hits_per_query,
error_on_multiple_hits=True)
# when a query hits multiple hits and error_on_multiple_hits=False
# the query should show up in multiple clusters
actual = clusters_from_uc_file(self.uc_lines_w_multiple_hits_per_query,
error_on_multiple_hits=False)
expected_clusters = {'s2': ['s2', 's3'],
's4': ['s4', 's3']}
expected_failures = ['s1']
expected_new_seeds = ['s2', 's4']
self.assertEqual(actual,
(expected_clusters, expected_failures, expected_new_seeds))
def generate_biom_table(seqs_fp, uc_fp, delim='_'):
"""Generate BIOM table and representative FASTA set
Parameters
----------
seqs_fp: string
file path to deblurred sequences
uc_fp: string
file path to dereplicated sequences map (.uc format)
delim: string, optional
delimiter for splitting sample and sequence IDs in sequence label
default: '_'
Returns
-------
deblur_clusters: dictionary
dictionary of clusters including dereplicated sequence labels
Table: biom.table
an instance of a BIOM table
"""
# parse clusters in dereplicated sequences map (.uc format)
with open(uc_fp, 'U') as uc_f:
derep_clusters, failures, seeds = clusters_from_uc_file(uc_f)
# parse clusters in deblur file, set observation ID to be the sequence
deblur_clusters = parse_deblur_output(seqs_fp, derep_clusters)
# create sparse dictionary of observation and sample ID counts
data, otu_ids, sample_ids = generate_biom_data(deblur_clusters, delim)
# build BIOM table
return deblur_clusters, Table(data,
otu_ids,
sample_ids,
observation_metadata=None,
sample_metadata=None,
table_id=None,
generated_by="deblur",
create_date=datetime.now().isoformat())
