def do(contigs_fpaths, output_dir, logger):
logger.print_timestamp()
logger.info('Running BUSCO...')
compilation_success = True
augustus_dirpath = download_augustus(logger)
if not augustus_dirpath:
compilation_success = False
elif not compile_tool('Augustus',
augustus_dirpath, [join('bin', 'augustus')],
logger=logger):
compilation_success = False
if compilation_success and not download_blast_binaries(
logger=logger, filenames=blast_filenames):
compilation_success = False
if not compilation_success:
logger.info('Failed finding conservative genes.')
return
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
tmp_dir = join(output_dir, 'tmp')
if not os.path.isdir(tmp_dir):
os.makedirs(tmp_dir)
n_jobs = min(len(contigs_fpaths), qconfig.max_threads)
busco_threads = max(1, qconfig.max_threads // n_jobs)
clade_dirpath = download_db(logger,
is_prokaryote=qconfig.prokaryote,
is_fungus=qconfig.is_fungus)
if not clade_dirpath:
logger.info('Failed finding conservative genes.')
return
config_fpath = make_config(output_dir, tmp_dir, busco_threads,
clade_dirpath, augustus_dirpath)
logger.info('Logs and results will be saved under ' + output_dir + '...')
os.environ['BUSCO_CONFIG_FILE'] = config_fpath
os.environ['AUGUSTUS_CONFIG_PATH'] = copy_augustus_configs(
augustus_dirpath, tmp_dir)
if not os.environ['AUGUSTUS_CONFIG_PATH']:
logger.error(
'Augustus configs not found, failed to run BUSCO without them.')
busco_args = [[
contigs_fpath,
qutils.label_from_fpath_for_fname(contigs_fpath)
] for contigs_fpath in contigs_fpaths]
summary_fpaths = run_parallel(busco_main_handler, busco_args,
qconfig.max_threads)
if not any(fpath for fpath in summary_fpaths):
logger.error(
'Failed running BUSCO for all the assemblies. See log files in ' +
output_dir + ' for information '
'(rerun with --debug to keep all intermediate files).')
return
# saving results
zero_output_for_all = True
for i, contigs_fpath in enumerate(contigs_fpaths):
report = reporting.get(contigs_fpath)
if summary_fpaths[i] and os.path.isfile(summary_fpaths[i]):
total_buscos, part_buscos, complete_buscos = 0, 0, 0
with open(summary_fpaths[i]) as f:
for line in f:
if 'Complete BUSCOs' in line:
complete_buscos = int(line.split()[0])
elif 'Fragmented' in line:
part_buscos = int(line.split()[0])
elif 'Total' in line:
total_buscos = int(line.split()[0])
if total_buscos != 0:
report.add_field(
reporting.Fields.BUSCO_COMPLETE,
('%.2f' % (float(complete_buscos) * 100.0 / total_buscos)))
report.add_field(reporting.Fields.BUSCO_PART,
('%.2f' %
(float(part_buscos) * 100.0 / total_buscos)))
if complete_buscos + part_buscos > 0:
zero_output_for_all = False
shutil.copy(summary_fpaths[i], output_dir)
else:
logger.error(
'Failed running BUSCO for ' + contigs_fpath +
'. See the log for detailed information'
' (rerun with --debug to keep all intermediate files).')
if zero_output_for_all:
logger.warning(
'BUSCO did not fail explicitly but found nothing for all assemblies! '
'Possible reasons and workarounds:\n'
' 1. Provided assemblies are so small that they do not contain even a single partial BUSCO gene. Not likely but may happen -- nothing to worry then.\n'
' 2. Incorrect lineage database was used. To run with fungi DB use --fungus, to run with eukaryota DB use --eukaryote, otherwise BUSCO uses bacteria DB.\n'
' 3. Problem with BUSCO dependencies, most likely Augustus. Check that the binaries in '
+ augustus_dirpath + '/bin/ are working properly.\n'
' If something is wrong with Augustus, you may try to install it yourself (https://github.com/Gaius-Augustus/Augustus) and add "augustus" binary to PATH.\n'
' 4. Some other problem with BUSCO. Check the logs (you may need to rerun QUAST with --debug to see all intermediate files).\n'
' If you cannot solve the problem yourself, post an issue at https://github.com/ablab/quast/issues or write to [email protected]'
)
if not qconfig.debug:
cleanup(output_dir)
logger.info('Done.')
def process_blast(blast_assemblies, downloaded_dirpath, corrected_dirpath,
labels, blast_check_fpath, err_fpath):
if not download_blast_binaries(filenames=blast_filenames):
return None, None, None
if qconfig.custom_blast_db_fpath:
global db_fpath
db_fpath = qconfig.custom_blast_db_fpath
if isdir(db_fpath):
db_aux_files = [
f for f in os.listdir(db_fpath) if f.endswith('.nsq')
]
if db_aux_files:
db_fpath = join(qconfig.custom_blast_db_fpath,
db_aux_files[0].replace('.nsq', ''))
elif isfile(db_fpath) and db_fpath.endswith('.nsq'):
db_fpath = db_fpath[:-len('.nsq')]
if not os.path.isfile(db_fpath + '.nsq'):
logger.error(
'You should specify path to BLAST database obtained by running makeblastdb command: '
'either path to directory containing <dbname>.nsq file or path to <dbname>.nsq file itself.'
' Also you can rerun MetaQUAST without --blast-db option. MetaQUAST uses SILVA 16S RNA database by default.',
exit_with_code=2)
elif not download_blastdb():
return None, None, None
blast_res_fpath = os.path.join(downloaded_dirpath, 'blast.res')
if len(blast_assemblies) > 0:
logger.main_info('Running BlastN..')
n_jobs = min(qconfig.max_threads, len(blast_assemblies))
blast_threads = max(1, qconfig.max_threads // n_jobs)
if is_python2():
from joblib2 import Parallel, delayed
else:
from joblib3 import Parallel, delayed
Parallel(n_jobs=n_jobs)(delayed(parallel_blast)(
assembly.fpath, assembly.label, corrected_dirpath, err_fpath,
blast_res_fpath, blast_check_fpath, blast_threads)
for i, assembly in enumerate(blast_assemblies))
logger.main_info()
species_scores = []
species_by_assembly = dict()
max_entries = 4
replacement_dict = defaultdict(list)
for label in labels:
assembly_scores = []
assembly_species = []
res_fpath = get_blast_output_fpath(blast_res_fpath, label)
if os.path.exists(res_fpath):
refs_for_query = 0
with open(res_fpath) as res_file:
query_id_col, subj_id_col, idy_col, len_col, score_col = None, None, None, None, None
for line in res_file:
fs = line.split()
if line.startswith('#'):
refs_for_query = 0
# Fields: query id, subject id, % identity, alignment length, mismatches, gap opens, q. start, q. end, s. start, s. end, evalue, bit score
if 'Fields' in line:
fs = line.strip().split('Fields: ')[-1].split(', ')
query_id_col = fs.index(
'query id') if 'query id' in fs else 0
subj_id_col = fs.index(
'subject id') if 'subject id' in fs else 1
idy_col = fs.index(
'% identity') if '% identity' in fs else 2
len_col = fs.index(
'alignment length'
) if 'alignment length' in fs else 3
score_col = fs.index(
'bit score') if 'bit score' in fs else 11
elif refs_for_query < max_entries and len(fs) > score_col:
query_id = fs[query_id_col]
organism_id = fs[subj_id_col]
idy = float(fs[idy_col])
length = int(fs[len_col])
score = float(fs[score_col])
if idy >= qconfig.identity_threshold and length >= qconfig.min_length and score >= qconfig.min_bitscore: # and (not scores or min(scores) - score < max_identity_difference):
seqname, taxons = parse_organism_id(organism_id)
if not seqname:
continue
species_name = get_species_name(seqname)
if species_name and 'uncultured' not in seqname:
if refs_for_query == 0:
if species_name not in assembly_species:
assembly_scores.append(
(seqname, query_id, score))
if taxons:
taxons_for_krona[correct_name(
seqname)] = taxons
assembly_species.append(species_name)
refs_for_query += 1
else:
seq_scores = [
(query_name, seq_query_id,
seq_score)
for query_name, seq_query_id,
seq_score in assembly_scores
if get_species_name(
query_name) == species_name
]
if seq_scores and score > seq_scores[
0][2]:
assembly_scores.remove(
seq_scores[0])
assembly_scores.append(
(seqname, query_id, score))
if taxons:
taxons_for_krona[correct_name(
seqname)] = taxons
refs_for_query += 1
else:
if seqname not in replacement_dict[
query_id]:
replacement_dict[query_id].append(
seqname)
refs_for_query += 1
assembly_scores = sorted(assembly_scores, reverse=True)
assembly_scores = assembly_scores[:qconfig.max_references]
for seqname, query_id, score in assembly_scores:
if not species_by_assembly or not any(
seqname in species_list
for species_list in species_by_assembly.values()):
species_scores.append((seqname, query_id, score))
species_by_assembly[label] = [
seqname for seqname, query_id, score in assembly_scores
]
if not species_scores:
return None, None, None
return species_scores, species_by_assembly, replacement_dict
def do(contigs_fpaths, output_dir, logger):
logger.print_timestamp()
logger.info('Running BUSCO...')
compilation_success = True
augustus_dirpath = download_augustus(logger)
if not augustus_dirpath:
compilation_success = False
elif not compile_tool('Augustus',
augustus_dirpath, [join('bin', 'augustus')],
logger=logger):
compilation_success = False
if compilation_success and not download_blast_binaries(
logger=logger, filenames=blast_filenames):
compilation_success = False
if not compilation_success:
logger.info('Failed finding conservative genes.')
return
set_augustus_dir(augustus_dirpath)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
tmp_dir = join(output_dir, 'tmp')
if not os.path.isdir(tmp_dir):
os.makedirs(tmp_dir)
n_jobs = min(len(contigs_fpaths), qconfig.max_threads)
busco_threads = max(1, qconfig.max_threads // n_jobs)
clade_dirpath = download_db(logger, is_prokaryote=qconfig.prokaryote)
if not clade_dirpath:
logger.info('Failed finding conservative genes.')
return
log_fpath = join(output_dir, 'busco.log')
logger.info('Logging to ' + log_fpath + '...')
busco_args = [([
'-i', contigs_fpath, '-o',
qutils.label_from_fpath_for_fname(contigs_fpath), '-l', clade_dirpath,
'-m', 'genome', '-f', '-z', '-c',
str(busco_threads), '-t', tmp_dir,
'--augustus_parameters=\'--AUGUSTUS_CONFIG_PATH=' +
join(augustus_dirpath, 'config') + '\''
], output_dir) for contigs_fpath in contigs_fpaths]
summary_fpaths = run_parallel(busco.main, busco_args, qconfig.max_threads)
if not any(fpath for fpath in summary_fpaths):
logger.error('Failed running BUSCO for all the assemblies. See ' +
log_fpath + ' for information.')
return
# saving results
for i, contigs_fpath in enumerate(contigs_fpaths):
report = reporting.get(contigs_fpath)
if summary_fpaths[i] and os.path.isfile(summary_fpaths[i]):
total_buscos, part_buscos, complete_buscos = 0, 0, 0
with open(summary_fpaths[i]) as f:
for line in f:
if 'Complete BUSCOs' in line:
complete_buscos = int(line.split()[0])
elif 'Fragmented' in line:
part_buscos = int(line.split()[0])
elif 'Total' in line:
total_buscos = int(line.split()[0])
if total_buscos != 0:
report.add_field(
reporting.Fields.BUSCO_COMPLETE,
('%.2f' % (float(complete_buscos) * 100.0 / total_buscos)))
report.add_field(reporting.Fields.BUSCO_PART,
('%.2f' %
(float(part_buscos) * 100.0 / total_buscos)))
else:
logger.error('Failed running BUSCO for ' + contigs_fpath +
'. See ' + log_fpath + ' for information.')
logger.info('Done.')
def do(contigs_fpaths, output_dir, logger):
logger.print_timestamp()
logger.info('Running BUSCO...')
compilation_success = True
augustus_dirpath = download_augustus(logger)
if not augustus_dirpath:
compilation_success = False
elif not compile_tool('Augustus',
augustus_dirpath, [join('bin', 'augustus')],
logger=logger):
compilation_success = False
if compilation_success and not download_blast_binaries(
logger=logger, filenames=blast_filenames):
compilation_success = False
if not compilation_success:
logger.info('Failed finding conservative genes.')
return
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
tmp_dir = join(output_dir, 'tmp')
if not os.path.isdir(tmp_dir):
os.makedirs(tmp_dir)
n_jobs = min(len(contigs_fpaths), qconfig.max_threads)
busco_threads = max(1, qconfig.max_threads // n_jobs)
clade_dirpath = download_db(logger,
is_prokaryote=qconfig.prokaryote,
is_fungus=qconfig.is_fungus)
if not clade_dirpath:
logger.info('Failed finding conservative genes.')
return
config_fpath = make_config(output_dir, tmp_dir, busco_threads,
clade_dirpath, augustus_dirpath)
logger.info('Logs and results will be saved under ' + output_dir + '...')
os.environ['BUSCO_CONFIG_FILE'] = config_fpath
os.environ['AUGUSTUS_CONFIG_PATH'] = copy_augustus_contigs(
augustus_dirpath, tmp_dir)
if not os.environ['AUGUSTUS_CONFIG_PATH']:
logger.error(
'Augustus configs not found, failed to run BUSCO without them.')
busco_args = [[
contigs_fpath,
qutils.label_from_fpath_for_fname(contigs_fpath)
] for contigs_fpath in contigs_fpaths]
summary_fpaths = run_parallel(busco_main_handler, busco_args,
qconfig.max_threads)
if not any(fpath for fpath in summary_fpaths):
logger.error(
'Failed running BUSCO for all the assemblies. See log files in ' +
output_dir + ' for information.')
return
# saving results
for i, contigs_fpath in enumerate(contigs_fpaths):
report = reporting.get(contigs_fpath)
if summary_fpaths[i] and os.path.isfile(summary_fpaths[i]):
total_buscos, part_buscos, complete_buscos = 0, 0, 0
with open(summary_fpaths[i]) as f:
for line in f:
if 'Complete BUSCOs' in line:
complete_buscos = int(line.split()[0])
elif 'Fragmented' in line:
part_buscos = int(line.split()[0])
elif 'Total' in line:
total_buscos = int(line.split()[0])
if total_buscos != 0:
report.add_field(
reporting.Fields.BUSCO_COMPLETE,
('%.2f' % (float(complete_buscos) * 100.0 / total_buscos)))
report.add_field(reporting.Fields.BUSCO_PART,
('%.2f' %
(float(part_buscos) * 100.0 / total_buscos)))
shutil.copy(summary_fpaths[i], output_dir)
else:
logger.error('Failed running BUSCO for ' + contigs_fpath +
'. See the log for detailed information.')
if not qconfig.debug:
cleanup(output_dir)
logger.info('Done.')
def do(contigs_fpaths, output_dir, logger):
logger.print_timestamp()
logger.info('Running BUSCO...')
compilation_success = True
augustus_dirpath = download_augustus(logger)
if not augustus_dirpath:
compilation_success = False
elif not compile_tool('Augustus', augustus_dirpath, [join('bin', 'augustus')], logger=logger):
compilation_success = False
if compilation_success and not download_blast_binaries(logger=logger, filenames=blast_filenames):
compilation_success = False
if not compilation_success:
logger.info('Failed finding conservative genes.')
return
set_augustus_dir(augustus_dirpath)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
tmp_dir = join(output_dir, 'tmp')
if not os.path.isdir(tmp_dir):
os.makedirs(tmp_dir)
n_jobs = min(len(contigs_fpaths), qconfig.max_threads)
busco_threads = max(1, qconfig.max_threads // n_jobs)
clade_dirpath = download_db(logger, is_prokaryote=qconfig.prokaryote, is_fungus=qconfig.is_fungus)
if not clade_dirpath:
logger.info('Failed finding conservative genes.')
return
log_fpath = join(output_dir, 'busco.log')
logger.info('Logging to ' + log_fpath + '...')
busco_args = [(['-i', contigs_fpath, '-o', qutils.label_from_fpath_for_fname(contigs_fpath), '-l', clade_dirpath,
'-m', 'genome', '-f', '-z', '-c', str(busco_threads), '-t', tmp_dir,
'--augustus_parameters=\'--AUGUSTUS_CONFIG_PATH=' + join(augustus_dirpath, 'config') + '\'' ], output_dir)
for contigs_fpath in contigs_fpaths]
summary_fpaths = run_parallel(busco.main, busco_args, qconfig.max_threads)
if not any(fpath for fpath in summary_fpaths):
logger.error('Failed running BUSCO for all the assemblies. See ' + log_fpath + ' for information.')
return
# saving results
for i, contigs_fpath in enumerate(contigs_fpaths):
report = reporting.get(contigs_fpath)
if summary_fpaths[i] and os.path.isfile(summary_fpaths[i]):
total_buscos, part_buscos, complete_buscos = 0, 0, 0
with open(summary_fpaths[i]) as f:
for line in f:
if 'Complete BUSCOs' in line:
complete_buscos = int(line.split()[0])
elif 'Fragmented' in line:
part_buscos = int(line.split()[0])
elif 'Total' in line:
total_buscos = int(line.split()[0])
if total_buscos != 0:
report.add_field(reporting.Fields.BUSCO_COMPLETE, ('%.2f' % (float(complete_buscos) * 100.0 / total_buscos)))
report.add_field(reporting.Fields.BUSCO_PART, ('%.2f' % (float(part_buscos) * 100.0 / total_buscos)))
else:
logger.error(
'Failed running BUSCO for ' + contigs_fpath + '. See ' + log_fpath + ' for information.')
logger.info('Done.')
def process_blast(blast_assemblies, downloaded_dirpath, corrected_dirpath, labels, blast_check_fpath, err_fpath):
if not download_blast_binaries(filenames=blast_filenames):
return None, None
if qconfig.custom_blast_db_fpath:
global db_fpath
db_fpath = qconfig.custom_blast_db_fpath
if isdir(db_fpath):
db_aux_files = [f for f in os.listdir(db_fpath) if f.endswith('.nsq')]
if db_aux_files:
db_fpath = join(qconfig.custom_blast_db_fpath, db_aux_files[0].replace('.nsq', ''))
elif isfile(db_fpath) and db_fpath.endswith('.nsq'):
db_fpath = db_fpath[:-len('.nsq')]
if not os.path.isfile(db_fpath + '.nsq'):
logger.error('You should specify path to BLAST database obtained by running makeblastdb command: '
'either path to directory containing <dbname>.nsq file or path to <dbname>.nsq file itself.'
' Also you can rerun MetaQUAST without --blast-db option. MetaQUAST uses SILVA 16S RNA database by default.',
exit_with_code=2)
elif not download_blastdb():
return None, None
blast_res_fpath = os.path.join(downloaded_dirpath, 'blast.res')
if len(blast_assemblies) > 0:
logger.main_info('Running BlastN..')
n_jobs = min(qconfig.max_threads, len(blast_assemblies))
blast_threads = max(1, qconfig.max_threads // n_jobs)
if is_python2():
from joblib import Parallel, delayed
else:
from joblib3 import Parallel, delayed
Parallel(n_jobs=n_jobs)(delayed(parallel_blast)(assembly.fpath, assembly.label, corrected_dirpath,
err_fpath, blast_res_fpath, blast_check_fpath, blast_threads)
for i, assembly in enumerate(blast_assemblies))
logger.main_info()
species_scores = []
species_by_assembly = dict()
max_entries = 4
replacement_dict = defaultdict(list)
for label in labels:
assembly_scores = []
assembly_species = []
res_fpath = get_blast_output_fpath(blast_res_fpath, label)
if os.path.exists(res_fpath):
refs_for_query = 0
with open(res_fpath) as res_file:
query_id_col, subj_id_col, idy_col, len_col, score_col = None, None, None, None, None
for line in res_file:
fs = line.split()
if line.startswith('#'):
refs_for_query = 0
# Fields: query id, subject id, % identity, alignment length, mismatches, gap opens, q. start, q. end, s. start, s. end, evalue, bit score
if 'Fields' in line:
fs = line.strip().split('Fields: ')[-1].split(', ')
query_id_col = fs.index('query id')
subj_id_col = fs.index('subject id')
idy_col = fs.index('% identity')
len_col = fs.index('alignment length')
score_col = fs.index('bit score')
elif refs_for_query < max_entries and len(fs) > score_col:
query_id = fs[query_id_col]
organism_id = fs[subj_id_col]
idy = float(fs[idy_col])
length = int(fs[len_col])
score = float(fs[score_col])
if idy >= qconfig.identity_threshold and length >= qconfig.min_length and score >= qconfig.min_bitscore: # and (not scores or min(scores) - score < max_identity_difference):
seqname, taxons = parse_organism_id(organism_id)
if not seqname:
continue
species_name = seqname.split('_')
if len(species_name) > 1 and 'uncultured' not in seqname:
species_name = species_name[0] + '_' + species_name[1]
if refs_for_query == 0:
if species_name not in assembly_species:
assembly_scores.append((seqname, query_id, score))
if taxons:
taxons_for_krona[correct_name(seqname)] = taxons
assembly_species.append(species_name)
refs_for_query += 1
else:
seq_scores = [(seqname, query_id, score) for seqname, query_id, score in assembly_scores
if species_name in seqname]
if seq_scores and score > seq_scores[0][2]:
assembly_scores.remove(seq_scores[0])
assembly_scores.append((seqname, query_id, score))
if taxons:
taxons_for_krona[correct_name(seqname)] = taxons
refs_for_query += 1
else:
if seqname not in replacement_dict[query_id]:
replacement_dict[query_id].append(seqname)
refs_for_query += 1
assembly_scores = sorted(assembly_scores, reverse=True)
assembly_scores = assembly_scores[:qconfig.max_references]
for seqname, query_id, score in assembly_scores:
if not species_by_assembly or not any(seqname in species_list for species_list in species_by_assembly.values()):
species_scores.append((seqname, query_id, score))
species_by_assembly[label] = [seqname for seqname, query_id, score in assembly_scores]
if not species_scores:
return None, None
return species_scores, species_by_assembly, replacement_dict
def process_blast(blast_assemblies, downloaded_dirpath, corrected_dirpath,
labels, blast_check_fpath, err_fpath):
if not download_blast_binaries(filenames=blast_filenames):
return None, None
if qconfig.custom_blast_db_fpath:
global db_fpath
db_fpath = qconfig.custom_blast_db_fpath
if isdir(db_fpath):
db_aux_files = [
f for f in os.listdir(db_fpath) if f.endswith('.nsq')
]
if db_aux_files:
db_fpath = join(qconfig.custom_blast_db_fpath,
db_aux_files[0].replace('.nsq', ''))
elif isfile(db_fpath) and db_fpath.endswith('.nsq'):
db_fpath = db_fpath[:-len('.nsq')]
if not os.path.isfile(db_fpath + '.nsq'):
logger.error(
'You should specify path to BLAST database obtained by running makeblastdb command: '
'either path to directory containing <dbname>.nsq file or path to <dbname>.nsq file itself.'
' Also you can rerun MetaQUAST without --blast-db option. MetaQUAST uses SILVA 16S RNA database by default.',
exit_with_code=2)
elif not download_blastdb():
return None, None
blast_res_fpath = os.path.join(downloaded_dirpath, 'blast.res')
if len(blast_assemblies) > 0:
logger.main_info('Running BlastN..')
n_jobs = min(qconfig.max_threads, len(blast_assemblies))
blast_threads = max(1, qconfig.max_threads // n_jobs)
if is_python2():
from joblib import Parallel, delayed
else:
from joblib3 import Parallel, delayed
Parallel(n_jobs=n_jobs)(delayed(parallel_blast)(
assembly.fpath, assembly.label, corrected_dirpath, err_fpath,
blast_res_fpath, blast_check_fpath, blast_threads)
for i, assembly in enumerate(blast_assemblies))
logger.main_info('')
scores_organisms = []
organisms_assemblies = {}
for label in labels:
all_scores = []
organisms = []
res_fpath = get_blast_output_fpath(blast_res_fpath, label)
if os.path.exists(res_fpath):
refs_for_query = 0
with open(res_fpath) as res_file:
for line in res_file:
if refs_for_query == 0 and not line.startswith(
'#') and len(line.split()) > 10:
# TODO: find and parse "Fields" line to detect each column indexes:
# Fields: query id, subject id, % identity, alignment length, mismatches, gap opens, q. start, q. end, s. start, s. end, evalue, bit score
# We need: identity, legnth, score, query and subject id.
line = line.split()
organism_id = line[1]
idy = float(line[2])
length = int(line[3])
score = float(line[11])
if idy >= qconfig.identity_threshold and length >= qconfig.min_length and score >= qconfig.min_bitscore: # and (not scores or min(scores) - score < max_identity_difference):
seqname, taxons = parse_organism_id(organism_id)
if not seqname:
continue
specie = seqname.split('_')
if len(specie) > 1 and 'uncultured' not in seqname:
specie = specie[0] + '_' + specie[1]
if specie not in organisms:
all_scores.append((score, seqname))
if taxons:
taxons_for_krona[correct_name(
seqname)] = taxons
organisms.append(specie)
refs_for_query += 1
else:
tuple_scores = [
x for x in all_scores if specie in x[1]
]
if tuple_scores and score > tuple_scores[
0][0]:
all_scores.remove((tuple_scores[0][0],
tuple_scores[0][1]))
all_scores.append((score, seqname))
if taxons:
taxons_for_krona[correct_name(
seqname)] = taxons
refs_for_query += 1
elif line.startswith('#'):
refs_for_query = 0
all_scores = sorted(all_scores, reverse=True)
all_scores = all_scores[:qconfig.max_references]
for score in all_scores:
if not organisms_assemblies or (
organisms_assemblies.values() and not [
1 for list in organisms_assemblies.values()
if score[1] in list
]):
scores_organisms.append(score)
organisms_assemblies[label] = [score[1] for score in all_scores]
if not scores_organisms:
return None, None
return scores_organisms, organisms_assemblies