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
def add_clusters_to_refpkg(refpkg, **kwargs):
with refpkg.open_resource('taxonomy') as tax_fp:
tax = taxtable.read(tax_fp)
with refpkg.open_resource('seq_info') as sinfo_fp:
reader = csv.DictReader(sinfo_fp)
sinfo = list(reader)
# Annotate
add_cluster_taxids(tax, sinfo, **kwargs)
with util.ntf(prefix='seq_info-', suffix='.csv') as seqinfo_tf, \
util.ntf(prefix='taxonomy-', suffix='.csv') as tax_tf:
w = csv.DictWriter(seqinfo_tf, reader.fieldnames)
w.writeheader()
w.writerows(sinfo)
seqinfo_tf.close()
tax.write_taxtable(tax_tf)
tax_tf.close()
refpkg.start_transaction()
refpkg.update_file('seq_info', seqinfo_tf.name)
refpkg.update_file('taxonomy', tax_tf.name)
refpkg.commit_transaction()