def main():
# Check version
if sys.version < '3.3':
your_version = sys.version.split(' ')[0]
print('* Your Python version (%s) is too old! Please upgrade to 3.3+!' % your_version)
sys.exit()
proglog = logmsg.message(prog='msaparser', cmd=' '.join(sys.argv))
options, opt_others = config.get_configuration(os.path.dirname(os.path.abspath(__file__)))
options.output_directory = options.output_directory.rstrip('/')
options.source_directory = options.source_directory.rstrip('/')
if not os.path.exists(options.output_directory + '/html'):
os.makedirs(options.output_directory + '/html')
mainfile = options.output_directory + '/' + opt_others.get('output_files').get('main')
with open(mainfile, 'w') as fw:
for msg in proglog.start_message():
fw.write(msg)
fw.write('\n')
fw.flush()
parser.writeheader(fw)
cluinput = msaio.FileInput(options.source_directory)
proc_manager = Manager()
q_write = proc_manager.Queue()
proc = pool.Pool(processes=options.process_num)
while cluinput.files:
if len(cluinput.files) / 100 >= 1:
files = cluinput.files[0:100]
cluinput.files = cluinput.files[100:]
else:
files = cluinput.files[0:len(cluinput.files)]
cluinput.files = cluinput.files[len(cluinput.files):]
tasks = []
for root, filename in files:
tasks.append((root, filename, options, opt_others, q_write))
proc.starmap(begin_parse, tasks)
write_result(q_write, options.output_directory, mainfile)
with open(mainfile, 'a') as fw:
fw.write('\n')
for msg in proglog.end_message():
fw.write(msg)
fw.flush()
def main():
proglog = logmsg.message(prog='fetchfa', cmd=' '.join(sys.argv))
parser = argparse.ArgumentParser(description='fetchfa - Fetch fasta files from Entrez')
parser.add_argument('input_file', nargs='?')
parser.add_argument('-d', '--db', dest='database', default='protein',
help='database (default: protein)')
parser.add_argument('-q', '--query', dest='query_id',
help='accessions to be fetched. If this option is specifid, the script will use the values '
'to fetch data, and no input file is required to be handled.')
parser.add_argument('-o', '--output', dest='output', default='fetchfa_out_' + name.genid(),
help='output directory or file name. If this option is not specified, the script will generate '
'one with unique identifier at current directory.')
parser.add_argument('-l', '--log', dest='log_file',
help='log file name')
args = parser.parse_args()
if args.log_file is None:
fwlog = open(args.output + '.log', 'w')
else:
fwlog = open(args.log_file, 'w')
for i in proglog.start_message():
fwlog.write(i)
fwlog.flush()
Entrez.email = name.genid() + '@example.com'
if args.query_id is not None:
with open(args.output + '.fa', 'w') as fw, open(args.output + '.log', 'w') as fwlog:
handle = Entrez.efetch(db=args.database,
id=args.query_id,
rettype='fasta',
retmode='text')
fw.write(handle.read())
fw.flush()
fwlog.write('# Fetched sequences: ' + str(len(args.query_id.split(','))) + '\n')
fwlog.write('#\n')
for i in proglog.end_message():
fwlog.write(i)
fwlog.flush()
else:
if not os.path.exists(args.output):
os.makedirs(args.output)
with open(args.input_file, 'r') as fin:
query_num = 0
for line in fin:
if line.lstrip() == '' or line.lstrip()[0] in ('#', 'a'):
continue
query_num += 1
with open(os.path.abspath(args.output) + '/' + line.split('\t')[0] + '.fa', 'w') as fw:
alist = line.rstrip().split('\t')[1].split(',')
while len(alist) > 30:
alist_part = alist[0:30]
alist = alist[30:len(alist)]
handle = Entrez.efetch(db=args.database,
id=','.join(alist_part),
rettype='fasta',
retmode='text')
fw.write(handle.read())
fw.flush()
handle.close()
handle = Entrez.efetch(db=args.database,
id=','.join(alist),
rettype='fasta',
retmode='text')
fw.write(handle.read())
fw.flush()
handle.close()
fwlog.write('# Fetched queries: ' + str(query_num) + '\n')
fwlog.write('#\n')
for i in proglog.end_message():
fwlog.write(i)
fwlog.flush()
fwlog.close()
def main():
proglog = logmsg.message(prog='blast2accmap', cmd=' '.join(sys.argv))
parser = argparse.ArgumentParser(description='blast2accmap - Extract names of query and hit sequences')
parser.add_argument('input_file')
parser.add_argument('-e', '--evalue', dest='ev_thresh', type=float, default=0.01,
help='evalue thresh (default: 0.01)')
parser.add_argument('-t', '--min_hit_num', dest='min_hit_num', type=int, default=1,
help='minimum number of hit sequences (default: 1)')
parser.add_argument('-o', '--output', dest='output_file',
help='output file name. If this option is not specified, the script will generate '
'one with unique identifier at current directory.')
args = parser.parse_args()
if args.output_file is None:
args.output_file = args.input_file + '_out_' + name.genid() + 'blastaccmap'
total_query_num = 0
parsed_query_num = 0
with open(args.input_file, 'r') as result_handle, open(args.output_file, 'w') as fw:
blast_records = NCBIXML.parse(result_handle)
for i in proglog.start_message():
fw.write(i)
fw.write('#\n')
fw.write('# E-value threshold: ' + str(args.ev_thresh) + '\n')
fw.write('# min hit number: ' + str(args.min_hit_num) + '\n')
fw.write('#\n')
fw.write('# filename query_accession,hit_accession_1,hit_accession_2, ...\n\n')
fw.flush()
gi = re.compile('gi\|(\d+)\|')
for blast_record in blast_records:
total_query_num += 1
if len(blast_record.alignments) < args.min_hit_num:
continue
hit_accs = []
for alignment in blast_record.alignments:
for hsp in alignment.hsps:
if alignment.accession in blast_record.query:
"""If query hit itself, ignore it. """
continue
if hsp.expect <= args.ev_thresh:
match = gi.match(alignment.hit_id).group(1)
if match is None:
print(alignment.accession + ' does not have gi.')
hit_accs.append(alignment.accession)
else:
hit_accs.append(match)
break
if len(hit_accs) >= args.min_hit_num:
parsed_query_num += 1
fw.write(blast_record.query + '\t' + blast_record.query + ',')
fw.write(','.join(hit_accs) + '\n')
fw.flush()
fw.write('\n')
fw.write('# Total queries: ' + str(total_query_num) + '\n')
fw.write('# Parsed queries: ' + str(parsed_query_num) + '\n')
fw.write('#\n')
for i in proglog.end_message():
fw.write(i)
fw.flush()
def main():
proglog = logmsg.message(prog='commonfa', cmd=' '.join(sys.argv))
parser = argparse.ArgumentParser(description='commonfa - Generate fasta files of sequences with common hit')
parser.add_argument('-b', '--blastlist', dest='input_files_blastlist', nargs='*', required=True,
help='blastlist files (required)')
parser.add_argument('-f', '--fasta', nargs='*', dest='input_files_fasta', required=True,
help='fasta files (required)')
parser.add_argument('-o', '--output-directory', dest='output', default='commonfa_out_' + name.genid(),
help='output directory. If this option is not specified, the script will generate '
'one with unique identifier at current directory.')
parser.add_argument('-p', '--process', dest='process_num', type=int, default=1,
help='number of threads (CPUs) to use')
args = parser.parse_args()
config = ConfigParser.ConfigParser()
config.read(os.path.dirname(os.path.abspath(__file__)) + '/config/group.cfg')
if not os.path.exists(args.output.rstrip('/') + '/msainput'):
os.makedirs(args.output.rstrip('/') + '/msainput')
fwlog = open(args.output.rstrip('.') + '/commonfa.log', 'w')
for i in proglog.start_message():
fwlog.write(i)
fwlog.flush()
awk_cmd = "awk -F'\t' '$5 ~ /ref/ { print $0 }' " + ' '.join(args.input_files_blastlist)
sort_cmd = "sort -t$'\t' -k5d,5 -k18g,18 -k22gr,22 -k19gr,19 -k26gr,26 -k6gr"
fwsort = open(args.output.rstrip('/') + '/sort.temp', 'w')
awk_proc = Popen(awk_cmd, stdout=PIPE, executable='/bin/bash', shell=True)
sort_proc = Popen(sort_cmd, stdin=awk_proc.stdout, stdout=fwsort, executable='/bin/bash', shell=True)
sort_proc.communicate()
fwsort.close()
fasta = {}
for filename in args.input_files_fasta:
fasta.update(dict(SeqIO.index(filename, 'fasta')))
susp_names = config.get('Susp', 'bdor').split(',')
res_names = config.get('Res', 'bdor').split(',')
rec_names = config.get('Rec', 'bdor').split(',')
has_susp = has_res = has_rec = False
commonhit = {}
hitname = re.compile('.*gi\|\d*?\|(.*?)\|(.*?)\|.*')
with open(args.output.rstrip('/') + '/sort.temp', 'r') as fin:
for line in fin:
data = line.split('\t')
match = hitname.match(data[4])
query_name = data[3]
hit_name = match.group(2)
query_frame = int(data[9])
if hit_name in commonhit:
if any(i in query_name for i in susp_names):
if has_susp is True:
continue
else:
has_susp = True
if any(i in query_name for i in res_names):
if has_res is True:
continue
else:
has_res = True
if any(i in query_name for i in rec_names):
if has_rec is True:
continue
else:
has_rec = True
commonhit[hit_name].append((query_name, query_frame))
else:
commonhit[hit_name] = [(query_name, query_frame)]
has_susp = has_res = has_rec = False
if any(i in query_name for i in susp_names):
has_susp = True
if any(i in query_name for i in res_names):
has_res = True
if any(i in query_name for i in rec_names):
has_rec = True
tasks = []
parsed_num = 0
for hit in commonhit:
if len(commonhit[hit]) == len(args.input_files_blastlist):
tasks.append((hit, commonhit[hit], fasta, args))
parsed_num += 1
pool = Pool(processes=args.process_num)
pool.map(do_parsing, tasks)
fwlog.write('# Parsed hits: ' + str(parsed_num) + '\n')
for i in proglog.end_message():
fwlog.write(i)
fwlog.flush()