if parts == header:
# Ignore the header line
continue
if not parts[-1] and len(parts) == len(header) + 1:
# Ignore dummy blank extra column, e.g.
# "...2.0\t\tPSORTb version 3.0\t\n"
parts = parts[:-1]
assert len(parts) == len(header), \
"%i fields, not %i, in line:\n%r" % (len(line), len(header), line)
out_handle.write(line)
count += 1
return count
# Note that if the input FASTA file contains no sequences,
# split_fasta returns an empty list (i.e. zero temp files).
fasta_files = split_fasta(fasta_file, os.path.join(tmp_dir, "tmhmm"), FASTA_CHUNK)
temp_files = [f + ".out" for f in fasta_files]
jobs = ["psort %s %s %s -o %s %s > %s" % (org_type, cutoff, divergent, out_type, fasta, temp)
for fasta, temp in zip(fasta_files, temp_files)]
def clean_up(file_list):
for f in file_list:
if os.path.isfile(f):
os.remove(f)
try:
os.rmdir(tmp_dir)
except Exception:
pass
if len(jobs) > 1 and num_threads > 1:
"Medium likely prediction",
"Highly likely prediction"]:
stop_err("ERROR: Problem with line: %r" % line)
out_handle.write("%s\t%s\t%s\t%s\n" % (identifier, position, score, likelihood))
return queries
working_dir, bin = get_path_and_binary()
if not os.path.isfile(fasta_file):
stop_err("ERROR: Missing input FASTA file %r" % fasta_file)
#Note that if the input FASTA file contains no sequences,
#split_fasta returns an empty list (i.e. zero temp files).
#We deliberately omit the FASTA descriptions to avoid a
#bug in promoter2 with descriptions over 200 characters.
fasta_files = split_fasta(fasta_file, os.path.join(tmp_dir, "promoter"), FASTA_CHUNK, keep_descr=False)
temp_files = [f+".out" for f in fasta_files]
jobs = ["%s %s > %s" % (bin, fasta, temp)
for fasta, temp in zip(fasta_files, temp_files)]
def clean_up(file_list):
for f in file_list:
if os.path.isfile(f):
os.remove(f)
try:
os.rmdir(tmp_dir)
except:
pass
if len(jobs) > 1 and num_threads > 1:
#A small "info" message for Galaxy to show the user.
)
)
tab_handle.close()
gff_handle.close()
if num_threads == 1:
# Still want to call split_fasta to apply truncation, but
# no reason to make multiple files - and more chance of
# hitting file system glitches if we do. So,
FASTA_CHUNK = sys.maxsize
fasta_files = split_fasta(
fasta_file,
os.path.join(tmp_dir, "signalp"),
n=FASTA_CHUNK,
truncate=truncate,
max_len=MAX_LEN,
)
temp_files = [f + ".out" for f in fasta_files]
assert len(fasta_files) == len(temp_files)
jobs = [
"signalp -short -t %s %s > %s" % (organism, fasta, temp)
for (fasta, temp) in zip(fasta_files, temp_files)
]
assert len(fasta_files) == len(temp_files) == len(jobs)
def clean_up(file_list):
"""Remove temp files, and if possible the temp directory."""
for f in file_list:
assert first60.startswith("First60="), line
first60 = first60[8:]
assert predhel.startswith("PredHel="), line
predhel = predhel[8:]
assert topology.startswith("Topology="), line
topology = topology[9:]
out_handle.write(
"%s\t%s\t%s\t%s\t%s\t%s\n" %
(identifier, length, exp_aa, first60, predhel, topology))
count += 1
return count
# Note that if the input FASTA file contains no sequences,
# split_fasta returns an empty list (i.e. zero temp files).
fasta_files = split_fasta(fasta_file, os.path.join(tmp_dir, "tmhmm"),
FASTA_CHUNK)
temp_files = [f + ".out" for f in fasta_files]
jobs = [
"tmhmm -short %s > %s" % (fasta, temp)
for fasta, temp in zip(fasta_files, temp_files)
]
def clean_up(file_list):
"""Remove temp files, and if possible the temp directory."""
for f in file_list:
if os.path.isfile(f):
os.remove(f)
try:
os.rmdir(tmp_dir)
except Exception:
num_threads = thread_count(sys.argv[2], default=4)
fasta_file = sys.argv[3]
tabular_file = sys.argv[4]
def clean_tabular(raw_handle, out_handle):
"""Clean up WoLF PSORT output to make it tabular."""
for line in raw_handle:
if not line or line.startswith("#"):
continue
name, data = line.rstrip("\r\n").split(None,1)
for rank, comp_data in enumerate(data.split(",")):
comp, score = comp_data.split()
out_handle.write("%s\t%s\t%s\t%i\n" \
% (name, comp, score, rank+1))
fasta_files = split_fasta(fasta_file, tabular_file, n=FASTA_CHUNK)
temp_files = [f+".out" for f in fasta_files]
assert len(fasta_files) == len(temp_files)
jobs = ["%s %s < %s > %s" % (exe, organism, fasta, temp)
for (fasta, temp) in zip(fasta_files, temp_files)]
assert len(fasta_files) == len(temp_files) == len(jobs)
def clean_up(file_list):
for f in file_list:
if os.path.isfile(f):
os.remove(f)
if len(jobs) > 1 and num_threads > 1:
#A small "info" message for Galaxy to show the user.
print "Using %i threads for %i tasks" % (min(num_threads, len(jobs)), len(jobs))
results = run_jobs(jobs, num_threads)
tabular_file = sys.argv[4]
def clean_tabular(raw_handle, out_handle):
"""Clean up WoLF PSORT output to make it tabular."""
for line in raw_handle:
if not line or line.startswith("#"):
continue
name, data = line.rstrip("\r\n").split(None, 1)
for rank, comp_data in enumerate(data.split(",")):
comp, score = comp_data.split()
out_handle.write("%s\t%s\t%s\t%i\n" %
(name, comp, score, rank + 1))
fasta_files = split_fasta(fasta_file, tabular_file, n=FASTA_CHUNK)
temp_files = [f + ".out" for f in fasta_files]
assert len(fasta_files) == len(temp_files)
jobs = [
"%s %s < %s > %s" % (exe, organism, fasta, temp)
for (fasta, temp) in zip(fasta_files, temp_files)
]
assert len(fasta_files) == len(temp_files) == len(jobs)
def clean_up(file_list):
for f in file_list:
if os.path.isfile(f):
os.remove(f)
)
)
tab_handle.close()
gff_handle.close()
if num_threads == 1:
# Still want to call split_fasta to apply truncation, but
# no reason to make multiple files - and more chance of
# hitting file system glitches if we do. So,
FASTA_CHUNK = sys.maxsize
fasta_files = split_fasta(
fasta_file,
os.path.join(tmp_dir, "signalp"),
n=FASTA_CHUNK,
truncate=truncate,
max_len=MAX_LEN,
)
temp_files = [f + ".out" for f in fasta_files]
assert len(fasta_files) == len(temp_files)
jobs = [
"signalp -short -t %s %s > %s" % (organism, fasta, temp)
for (fasta, temp) in zip(fasta_files, temp_files)
]
assert len(fasta_files) == len(temp_files) == len(jobs)
def clean_up(file_list):
"""Remove temp files, and if possible the temp directory."""
for f in file_list:
