def identify_otus_unnamed(seq_file, cluster_similarity):
"""
Generates sequence ids in a cluster
Identify sequences in OTUs at the given cluster similarity;
"""
logging.info("Running UCLUST on unnamed sequences at %f", cluster_similarity)
with util.ntf(prefix="uclust") as tf:
# Sort and cluster
uclust.cluster(seq_file, tf.name, pct_id=cluster_similarity, quiet=True)
clusters = uclust.sequences_by_cluster(uclust.parse_uclust_out(tf))
for _, sequences in clusters:
yield [i.query_label for i in sequences]
def identify_otus_unnamed(seq_file, cluster_similarity):
"""
Generates sequence ids in a cluster
Identify sequences in OTUs at the given cluster similarity;
"""
logging.info('Running UCLUST on unnamed sequences at %f',
cluster_similarity)
with util.ntf(prefix='uclust') as tf:
# Sort and cluster
uclust.cluster(
seq_file, tf.name, pct_id=cluster_similarity, quiet=True)
clusters = uclust.sequences_by_cluster(uclust.parse_uclust_out(tf))
for _, sequences in clusters:
yield [i.query_label for i in sequences]
def cluster(seqfile, seqnames, identity=1.0, prefix='cluster-', threads=None):
with util.ntf(prefix=prefix, suffix='.fasta') as fa, \
util.ntf(prefix=prefix, suffix='.uc') as uc:
wrap.esl_sfetch(seqfile, seqnames, fa)
fa.flush()
uclust.cluster(fa.name,
uc.name,
pct_id=identity,
pre_sorted=False,
quiet=True,
threads=threads)
df = uclust.parse_uclust_as_df(uc)
df = df[df.type != 'C']
df = df[['type', 'query_label', 'target_label']]
return df
def cluster(seqfile, seqnames, identity=1.0, prefix='cluster-', threads=None):
prefix = prefix.replace('/', '\\') # / confuses the filesystem
with util.ntf(prefix=prefix, suffix='.fasta') as fa, \
util.ntf(prefix=prefix, suffix='.uc') as uc:
wrap.esl_sfetch(seqfile, seqnames, fa)
fa.flush()
uclust.cluster(fa.name,
uc.name,
pct_id=identity,
pre_sorted=False,
quiet=True,
threads=threads)
df = uclust.parse_uclust_as_df(uc)
df = df[df.type != 'C']
df = df[['type', 'query_label', 'target_label']]
return df
