def process_nog_hits_file(hits_file,
query_fasta,
og2level,
skip_queries=None,
translate=False,
cpu=1,
excluded_taxa=None,
base_tempdir=None):
sequences = {
name: seq
for name, seq in seqio.iter_fasta_seqs(query_fasta,
translate=translate)
}
cmds = []
visited_queries = set()
if skip_queries:
visited_queries.update(skip_queries)
tempdir = mkdtemp(prefix='emappertmp_phmmer_', dir=base_tempdir)
for line in gopen(hits_file):
if line.startswith('#'):
continue
fields = map(str.strip, line.split('\t'))
seqname = fields[0]
if fields[1] == '-' or fields[1] == 'ERROR':
continue
if seqname in visited_queries:
continue
hitname = cleanup_og_name(fields[1])
level = og2level[hitname]
seq = sequences[seqname]
visited_queries.add(seqname)
target_fasta = os.path.join(get_fasta_path(), level, "%s.fa" % hitname)
cmds.append([seqname, seq, target_fasta, excluded_taxa, tempdir])
if cmds:
pool = multiprocessing.Pool(cpu)
for r in pool.imap(search.refine_hit, cmds):
yield r
pool.terminate()
shutil.rmtree(tempdir)
def process_nog_hits_file(hits_file, query_fasta, og2level, skip_queries=None,
translate=False, cpu=1, excluded_taxa=None):
sequences = {name: seq for name, seq in seqio.iter_fasta_seqs(
query_fasta, translate=translate)}
cmds = []
visited_queries = set()
if skip_queries:
visited_queries.update(skip_queries)
tempdir = mkdtemp(prefix='emappertmp_phmmer_', dir=TEMPDIR)
for line in gopen(hits_file):
if line.startswith('#'):
continue
fields = map(str.strip, line.split('\t'))
seqname = fields[0]
if fields[1] == '-' or fields[1] == 'ERROR':
continue
if seqname in visited_queries:
continue
hitname = cleanup_og_name(fields[1])
level = og2level[hitname]
seq = sequences[seqname]
visited_queries.add(seqname)
target_fasta = os.path.join(FASTA_PATH, level, "%s.fa" % hitname)
cmds.append([seqname, seq, target_fasta, excluded_taxa, tempdir])
if cmds:
pool = multiprocessing.Pool(cpu)
for r in pool.imap(search.refine_hit, cmds):
yield r
pool.terminate()
shutil.rmtree(tempdir)