def main():
parser = argparse.ArgumentParser(description='fixname - Fix hit name in the blastlist')
parser.add_argument('input_file')
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() + '.fix'
hitname = re.compile('.*?(gi\|\d*?\|.*?\|.*?\|)(.*)')
with open(args.input_file, 'r') as fin, open(args.output_file, 'w') as fw:
for linum, line in enumerate(fin, start=1):
if line.lstrip() == '' or line.lstrip()[0] in ('#', 'a'):
fw.write(line)
fw.flush()
else:
data = line.split('\t')
match = hitname.match(data[26])
if match is None:
print('No mathced name in line ' + str(linum) + '.')
print('Please have a check.')
sys.exit()
else:
data[4] = match.group(1)
data[26] = match.group(1) + match.group(2) + '\n'
fw.write('\t'.join(data))
fw.flush()
def main():
parser = argparse.ArgumentParser(description='fa2lens - Extract length data from a fasta file')
parser.add_argument('input_file')
parser.add_argument('-s', '--sep', dest='sep', default='\n',
help='seperator (default: newline)')
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() + '.leng.txt'
with open(args.input_file, 'r') as fin, open(args.output_file, 'w') as fw:
records = map(str, map(len, list(SeqIO.parse(fin, 'fasta'))))
fw.write(args.sep.join(records))
fw.flush()
def main():
parser = argparse.ArgumentParser(description='blastnol - Find the non-overlapping hits in the blast result')
parser.add_argument('input_file')
parser.add_argument('-o', '--output-directory', dest='output_dir',
help='output directory name. If this option is not specified, the script will generate '
'one with unique identifier at current directory.')
parser.add_argument('-q', '--query-sequence', dest='query_fa',
help='fasta file of query sequence. If this option is specified, the script will '
'generate a new fasta file that contains truncated long sequences.')
args = parser.parse_args()
if args.output_dir is None:
args.output_dir = args.input_file + '_out_' + name.genid()
else:
args.output_dir = args.output_dir.rstrip('/')
if not os.path.exists(args.output_dir):
os.makedirs(args.output_dir)
if args.query_fa is not None:
query_fa = dict(SeqIO.index(args.query_fa, 'fasta'))
fw_fa = open(args.output_dir + '/truncated.fa', 'w')
with open(args.output_dir + '/sort.temp', 'w') as fwsort:
awk_cmd = "awk -F'\t' 'int($1) { print $0 }' " + args.input_file
sort_cmd = "sort -t$'\t' -k9g,9 -k4d,4 -k18g,18 -k22gr,22 -k19gr,19 -k26gr,26 -k6gr"
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()
seq = {}
with open(args.output_dir + '/sort.temp', 'r') as fi:
for line in fi:
data = line.split('\t')
query_name = data[3]
hit_name = data[4]
query_strand = int(data[8])
if query_strand < 0:
query_name = '-' + query_name
query_hsp_start = int(data[7])
query_hsp_end = int(data[6])
else:
query_hsp_start = int(data[6])
query_hsp_end = int(data[7])
if query_name in seq:
for i in range(len(seq[query_name])):
if hit_name in seq[query_name][i][0]:
seq[query_name][i][1].append((query_hsp_start, query_hsp_end, [line]))
break
else:
seq[query_name].append((hit_name, [(query_hsp_start, query_hsp_end, [line])]))
else:
seq.update({query_name: [(hit_name, [(query_hsp_start, query_hsp_end, [line])])]})
# Combine hsps
for query_name in seq:
hit_rank = 1
for i in range(len(seq[query_name])):
hit = seq[query_name][i]
if len(hit[1]) > 1:
# Combine hsps
pos_start, pos_end, lines = combine_hsps(hit[1])
seq[query_name][i] = ([hit[0]], [(pos_start, pos_end)], [lines], [hit_rank])
else:
seq[query_name][i] = ([hit[0]], [(hit[1][0][0], hit[1][0][1])], [hit[1][0][2]], [hit_rank])
hit_rank += 1
# Sort hits by start position
for query_name in seq:
seq[query_name] = sorted(seq[query_name], key=get_start_pos)
# Check overlap
for query_name, hits in seq.items():
while len(hits) > 1:
position = calculate.get_non_overlap((hits[0][1][0][0], hits[0][1][-1][1]), (hits[1][1][0][0], hits[1][1][0][1]))
if position is not None:
# The two sequences are non-overlapping, combine them
seq[query_name][0] = (hits[0][0] + hits[1][0], hits[0][1] + hits[1][1], hits[0][2] + hits[1][2], hits[0][3] + hits[1][3])
seq[query_name].pop(1)
else:
# Compare the hit rank to determine which one is retained
if seq[query_name][0][3][-1] < seq[query_name][1][3][0]:
# Discard the next hit
seq[query_name].pop(1)
else:
# Discard the last hit, and join the next hit
seq[query_name][0][0].pop(-1)
seq[query_name][0][1].pop(-1)
seq[query_name][0][2].pop(-1)
seq[query_name][0][3].pop(-1)
seq[query_name][0] = (hits[0][0] + hits[1][0], hits[0][1] + hits[1][1], hits[0][2] + hits[1][2], hits[0][3] + hits[1][3])
seq[query_name].pop(1)
# Write data
with open(args.output_dir + '/hit_cover.tsv', 'w') as fw:
query_num = 0
query_num_cover_eq_two = 0
query_num_cover_eq_three = 0
query_num_cover_ge_four = 0
hit_set = set()
hr = header.blastlist()
fw.write(hr.get_all_tab() + '\n')
fw.flush()
for query, hits in seq.items():
query = query.split(' ')[0]
if len(hits[0][0]) > 1:
query_num += 1
if len(hits[0][0]) == 2:
query_num_cover_eq_two += 1
elif len(hits[0][0]) == 3:
query_num_cover_eq_three += 1
else:
query_num_cover_ge_four += 1
for lines in hits[0][2]:
for line in lines:
hit_set.add(line.split('\t')[4])
fw.write(line)
fw.flush()
if args.query_fa is not None:
# Truncated queries
segment_num = 0
for pos_start, pos_end in hits[0][1]:
fw_fa.write('>' + query + '_s' + str(segment_num) + '\n')
fw_fa.write(query_fa[query].seq.tostring()[pos_start - 1:pos_end] + '\n')
fw_fa.flush()
segment_num += 1
query_fa.pop(query)
else:
if args.query_fa is not None:
# Full-sequece queries
fw_fa.write('>' + query + '\n')
fw_fa.write(query_fa[query].seq.tostring() + '\n')
query_fa.pop(query)
if args.query_fa is not None:
# No-hit queries
for query in query_fa:
fw_fa.write('>' + query + '\n')
fw_fa.write(query_fa[query].seq.tostring() + '\n')
fw_fa.flush()
fw_fa.close()
fw.write('\n')
fw.write('# Number of queries that cover >= 2 hits: ' + str(query_num) + '\n')
fw.write('# Cover 2 hits: ' + str(query_num_cover_eq_two) + '\n')
fw.write('# Cover 3 hits: ' + str(query_num_cover_eq_three) + '\n')
fw.write('# Cover >= 4 hits: ' + str(query_num_cover_ge_four) + '\n')
fw.write('# Number of covered hits: ' + str(len(hit_set)))
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():
parser = argparse.ArgumentParser(description='commutate - Find the common mutation profile')
parser.add_argument('input', nargs='*')
parser.add_argument('-o', '--output-directory', dest='output', default='commutate_out_' + name.genid(),
help='output directory. If this option is not specified, the script will generate '
'one with unique identifier at current directory.')
args = parser.parse_args()
args.output = args.output.rstrip('/')
if not os.path.exists(args.output):
os.makedirs(args.output)
source = []
source_res_eq_susp = {}
source_rec_eq_susp = {}
source_rec_eq_res = {}
for afile in args.input:
source.append(afile)
source_res_eq_susp[afile] = {}
source_rec_eq_susp[afile] = {}
source_rec_eq_res[afile] = {}
with open(afile, 'r') as fin:
for line in fin:
if line.lstrip() == '' or line.lstrip()[0] in ('#', 'm'):
continue
data = line.rstrip().split('\t')
if int(data[3]) > 0:
source_res_eq_susp[afile].update({data[1]: set(data[9].split(','))})
if int(data[4]) > 0:
source_rec_eq_susp[afile].update({data[1]: set(data[10].split(','))})
if int(data[5]) > 0:
source_rec_eq_res[afile].update({data[1]: set(data[11].split(','))})
common_hitname_res_eq_susp = get_common_hitname(source_res_eq_susp)
common_mutation_profile_res_eq_susp = get_common_mutate(source_res_eq_susp, common_hitname_res_eq_susp)
common_hitname_rec_eq_susp = get_common_hitname(source_rec_eq_susp)
common_mutation_profile_rec_eq_susp = get_common_mutate(source_rec_eq_susp, common_hitname_rec_eq_susp)
common_hitname_rec_eq_res = get_common_hitname(source_rec_eq_res)
common_mutation_profile_rec_eq_res = get_common_mutate(source_rec_eq_res, common_hitname_rec_eq_res)
writefile(args.output + '/common_mutation_profile_res_eq_susp.txt', common_mutation_profile_res_eq_susp)
writefile(args.output + '/common_mutation_profile_rec_eq_susp.txt', common_mutation_profile_rec_eq_susp)
writefile(args.output + '/common_mutation_profile_rec_eq_res.txt', common_mutation_profile_rec_eq_res)
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()
