def update_existing(input_fasta,
tmp_dir,
ref_mash_db,
tmp_cluster_file,
header,
tmp_ref_fasta_file,
update_fasta,
num_threads=1):
sequences = read_fasta_dict(input_fasta)
for id in sequences:
seq = sequences[id]
tmp_fasta = os.path.join(tmp_dir, id + '_tmp.fasta')
with open(tmp_fasta, "w", encoding="utf-8") as fh:
fh.write("\n>{}\n{}\n".format(id, seq))
fh.close()
tmp_mash = os.path.join(tmp_dir, id + '_tmp.txt')
clust_assignments = add_new_record(tmp_fasta, ref_mash_db, tmp_mash,
tmp_cluster_file, (0.05, 0.0001),
num_threads)
writeClusterAssignments(tmp_cluster_file, header, clust_assignments)
with open(tmp_ref_fasta_file, "a", encoding="utf-8") as fh:
fh.write("\n>{}\n{}\n".format(id, seq))
fh.close()
clust_dict = selectCluster(clust_assignments, 1)
updateFastaFile(tmp_ref_fasta_file, update_fasta, clust_dict)
def main():
args = parse_args()
if args.debug:
init_console_logger(3)
logging.info("MOB-recon v. {} ".format(__version__))
if not args.outdir:
logging.error(
'Error, no output directory specified, please specify one')
sys.exit(-1)
if not args.infile:
logging.error('Error, no fasta specified, please specify one')
sys.exit(-1)
if not os.path.isfile(args.infile):
logging.error('Error, input fasta file does not exist: "{}"'.format(
args.infile))
sys.exit(-1)
logging.info('Processing fasta file {}'.format(args.infile))
logging.info('Analysis directory {}'.format(args.outdir))
if not os.path.isdir(args.outdir):
os.mkdir(args.outdir, 0o755)
# Check that the needed databases have been initialized
database_dir = os.path.abspath(args.database_directory)
verify_init(logging, database_dir)
status_file = os.path.join(database_dir, 'status.txt')
if not os.path.isfile(status_file):
logging.info(
'Warning! Needed databases have not been initialize please run mob_init and try again'
)
mob_suite.mob_init.main()
plasmid_files = dict()
input_fasta = args.infile
out_dir = args.outdir
num_threads = args.num_threads
tmp_dir = os.path.join(out_dir, '__tmp')
file_id = os.path.basename(input_fasta)
fixed_fasta = os.path.join(tmp_dir, 'fixed.input.fasta')
chromosome_file = os.path.join(out_dir, 'chromosome.fasta')
replicon_blast_results = os.path.join(tmp_dir,
'replicon_blast_results.txt')
mob_blast_results = os.path.join(tmp_dir, 'mobrecon_blast_results.txt')
repetitive_blast_results = os.path.join(tmp_dir,
'repetitive_blast_results.txt')
contig_blast_results = os.path.join(tmp_dir, 'contig_blast_results.txt')
# Input numeric params
min_rep_ident = float(args.min_rep_ident)
min_mob_ident = float(args.min_mob_ident)
min_con_ident = float(args.min_con_ident)
min_rpp_ident = float(args.min_rpp_ident)
idents = {
'min_rep_ident': min_rep_ident,
'min_mob_ident': min_mob_ident,
'min_con_ident': min_con_ident,
'min_rpp_ident': min_rpp_ident
}
for param in idents:
value = float(idents[param])
if value < 60:
logging.error(
"Error: {} is too low, please specify an integer between 70 - 100"
.format(param))
sys.exit(-1)
if value > 100:
logging.error(
"Error: {} is too high, please specify an integer between 70 - 100"
.format(param))
sys.exit(-1)
min_rep_cov = float(args.min_rep_cov)
min_mob_cov = float(args.min_mob_cov)
min_con_cov = float(args.min_con_cov)
min_rpp_cov = float(args.min_rpp_cov)
covs = {
'min_rep_cov': min_rep_cov,
'min_mob_cov': min_mob_cov,
'min_con_cov': min_con_cov,
'min_rpp_cov': min_rpp_cov
}
for param in covs:
value = float(covs[param])
if value < 60:
logging.error(
"Error: {} is too low, please specify an integer between 50 - 100"
.format(param))
sys.exit(-1)
if value > 100:
logging.error(
"Error: {} is too high, please specify an integer between 50 - 100"
.format(param))
sys.exit(-1)
min_rep_evalue = float(args.min_rep_evalue)
min_mob_evalue = float(args.min_mob_evalue)
min_con_evalue = float(args.min_con_evalue)
min_rpp_evalue = float(args.min_rpp_evalue)
evalues = {
'min_rep_evalue': min_rep_evalue,
'min_mob_evalue': min_mob_evalue,
'min_con_evalue': min_con_evalue,
'min_rpp_evalue': min_rpp_evalue
}
for param in evalues:
value = float(evalues[param])
if value > 1:
logging.error(
"Error: {} is too high, please specify an float evalue between 0 to 1"
.format(param))
sys.exit(-1)
min_overlapp = int(args.min_overlap)
min_length = int(args.min_length)
# Input numeric params
min_rep_ident = float(args.min_rep_ident)
min_mob_ident = float(args.min_mob_ident)
min_con_ident = float(args.min_con_ident)
min_rpp_ident = float(args.min_rpp_ident)
idents = {
'min_rep_ident': min_rep_ident,
'min_mob_ident': min_mob_ident,
'min_con_ident': min_con_ident,
'min_rpp_ident': min_rpp_ident
}
for param in idents:
value = idents[param]
if value < 60:
logging.error(
"Error: {} is too low, please specify an integer between 70 - 100"
.format(param))
sys.exit(-1)
if value > 100:
logging.error(
"Error: {} is too high, please specify an integer between 70 - 100"
.format(param))
sys.exit(-1)
min_rep_cov = float(args.min_rep_cov)
min_mob_cov = float(args.min_mob_cov)
min_con_cov = float(args.min_con_cov)
min_rpp_cov = float(args.min_rpp_cov)
covs = {
'min_rep_cov': min_rep_cov,
'min_mob_cov': min_mob_cov,
'min_con_cov': min_con_cov,
'min_rpp_cov': min_rpp_cov
}
for param in covs:
value = covs[param]
if value < 60:
logging.error(
"Error: {} is too low, please specify an integer between 50 - 100"
.format(param))
sys.exit(-1)
if value > 100:
logging.error(
"Error: {} is too high, please specify an integer between 50 - 100"
.format(param))
sys.exit(-1)
min_rep_evalue = float(args.min_rep_evalue)
min_mob_evalue = float(args.min_mob_evalue)
min_con_evalue = float(args.min_con_evalue)
min_rpp_evalue = float(args.min_rpp_evalue)
evalues = {
'min_rep_evalue': min_rep_evalue,
'min_mob_evalue': min_mob_evalue,
'min_con_evalue': min_con_evalue,
'min_rpp_evalue': min_rpp_evalue
}
for param in evalues:
value = evalues[param]
if value > 1:
logging.error(
"Error: {} is too high, please specify an float evalue between 0 to 1"
.format(param))
sys.exit(-1)
min_overlapp = args.min_overlap
min_length = args.min_length
# Input Databases
default_database_dir = os.path.join(
os.path.dirname(os.path.realpath(__file__)), 'databases')
if database_dir == default_database_dir:
plasmid_ref_db = args.plasmid_db
replicon_ref = args.plasmid_replicons
mob_ref = args.plasmid_mob
mash_db = args.plasmid_mash_db
repetitive_mask_file = args.repetitive_mask
else:
plasmid_ref_db = os.path.join(database_dir,
'ncbi_plasmid_full_seqs.fas')
replicon_ref = os.path.join(database_dir, 'rep.dna.fas')
mob_ref = os.path.join(database_dir, 'mob.proteins.faa')
mash_db = os.path.join(database_dir, 'ncbi_plasmid_full_seqs.fas.msh')
repetitive_mask_file = os.path.join(database_dir, 'repetitive.dna.fas')
check_dependencies(logging)
needed_dbs = [
plasmid_ref_db, replicon_ref, mob_ref, mash_db, repetitive_mask_file,
"{}.nin".format(repetitive_mask_file)
]
for db in needed_dbs:
if (not os.path.isfile(db)):
logging.error('Error needed database missing "{}"'.format(db))
sys.exit(-1)
contig_report_file = os.path.join(out_dir, 'contig_report.txt')
minimus_prefix = os.path.join(tmp_dir, 'minimus')
filtered_blast = os.path.join(tmp_dir, 'filtered_blast.txt')
repetitive_blast_report = os.path.join(out_dir,
'repetitive_blast_report.txt')
mobtyper_results_file = os.path.join(out_dir,
'mobtyper_aggregate_report.txt')
keep_tmp = args.keep_tmp
run_circlator = args.run_circlator
unicycler_contigs = args.unicycler_contigs
if not isinstance(args.num_threads, int):
logging.info(
'Error number of threads must be an integer, you specified "{}"'.
format(args.num_threads))
logging.info('Creating tmp working directory {}'.format(tmp_dir))
if not os.path.isdir(tmp_dir):
os.mkdir(tmp_dir, 0o755)
logging.info(
'Writing cleaned header input fasta file from {} to {}'.format(
input_fasta, fixed_fasta))
fix_fasta_header(input_fasta, fixed_fasta)
contig_seqs = read_fasta_dict(fixed_fasta)
logging.info('Running replicon blast on {}'.format(replicon_ref))
replicon_contigs = getRepliconContigs(
replicon_blast(replicon_ref,
fixed_fasta,
min_rep_ident,
min_rep_cov,
min_rep_evalue,
tmp_dir,
replicon_blast_results,
num_threads=num_threads))
logging.info('Running relaxase blast on {}'.format(mob_ref))
mob_contigs = getRepliconContigs(
mob_blast(mob_ref,
fixed_fasta,
min_mob_ident,
min_mob_cov,
min_mob_evalue,
tmp_dir,
mob_blast_results,
num_threads=num_threads))
logging.info('Running contig blast on {}'.format(plasmid_ref_db))
contig_blast(fixed_fasta, plasmid_ref_db, min_con_ident, min_con_cov,
min_con_evalue, min_length, tmp_dir, contig_blast_results)
pcl_clusters = contig_blast_group(filtered_blast, min_overlapp)
logging.info(
'Running repetitive contig masking blast on {}'.format(mob_ref))
repetitive_contigs = repetitive_blast(fixed_fasta,
repetitive_mask_file,
min_rpp_ident,
min_rpp_cov,
min_rpp_evalue,
min_length,
tmp_dir,
repetitive_blast_results,
num_threads=num_threads)
circular_contigs = dict()
logging.info('Running circlator minimus2 on {}'.format(fixed_fasta))
if run_circlator:
circular_contigs = circularize(fixed_fasta, minimus_prefix)
if unicycler_contigs:
for seqid in contig_seqs:
if 'circular=true' in seqid:
circular_contigs[seqid] = ''
repetitive_dna = dict()
results_fh = open(repetitive_blast_report, 'w')
results_fh.write(
"contig_id\tmatch_id\tmatch_type\tscore\tcontig_match_start\tcontig_match_end\n"
)
for contig_id in repetitive_contigs:
match_info = repetitive_contigs[contig_id]['id'].split('|')
repetitive_dna[contig_id] = "{}\t{}\t{}\t{}\t{}".format(
match_info[1], match_info[len(match_info) - 1],
repetitive_contigs[contig_id]['score'],
repetitive_contigs[contig_id]['contig_start'],
repetitive_contigs[contig_id]['contig_end'])
results_fh.write("{}\t{}\t{}\t{}\t{}\t{}\n".format(
contig_id, match_info[1], match_info[len(match_info) - 1],
repetitive_contigs[contig_id]['score'],
repetitive_contigs[contig_id]['contig_start'],
repetitive_contigs[contig_id]['contig_end']))
results_fh.close()
seq_clusters = dict()
cluster_bitscores = dict()
for seqid in pcl_clusters:
cluster_id = list(pcl_clusters[seqid].keys())[0]
bitscore = pcl_clusters[seqid][cluster_id]
cluster_bitscores[cluster_id] = bitscore
sorted_cluster_bitscores = sorted(list(cluster_bitscores.items()),
key=operator.itemgetter(1))
sorted_cluster_bitscores.reverse()
contigs_assigned = dict()
for cluster_id, bitscore in sorted_cluster_bitscores:
if not cluster_id in seq_clusters:
seq_clusters[cluster_id] = dict()
for seqid in pcl_clusters:
if not cluster_id in pcl_clusters[seqid]:
continue
if seqid in contig_seqs and seqid not in contigs_assigned:
seq_clusters[cluster_id][seqid] = contig_seqs[seqid]
contigs_assigned[seqid] = cluster_id
# Add sequences with known replicons regardless of whether they belong to a mcl cluster
clust_id = 0
refined_clusters = dict()
for contig_id in mob_contigs:
if not contig_id in pcl_clusters:
if contig_id in contig_seqs:
if not clust_id in seq_clusters:
seq_clusters["Novel_" + str(clust_id)] = dict()
if not contig_id in pcl_clusters:
pcl_clusters[contig_id] = dict()
pcl_clusters[contig_id]["Novel_" + str(clust_id)] = 0
seq_clusters["Novel_" +
str(clust_id)][contig_id] = contig_seqs[contig_id]
clust_id += 1
# Add sequences with known relaxases regardless of whether they belong to a mcl cluster
count_replicons = dict()
for contig_id in replicon_contigs:
if not contig_id in pcl_clusters:
if contig_id in contig_seqs:
if not clust_id in seq_clusters:
seq_clusters["Novel_" + str(clust_id)] = dict()
if not contig_id in pcl_clusters:
pcl_clusters[contig_id] = dict()
pcl_clusters[contig_id]["Novel_" + str(clust_id)] = dict()
seq_clusters["Novel_" +
str(clust_id)][contig_id] = contig_seqs[contig_id]
clust_id += 1
refined_clusters = dict()
# split out circular sequences from each other
replicon_clusters = dict()
for contig_id in replicon_contigs:
for hit_id in replicon_contigs[contig_id]:
id, rep_type = hit_id.split('|')
cluster = list(pcl_clusters[contig_id].keys())[0]
if not cluster in replicon_clusters:
replicon_clusters[cluster] = 0
replicon_clusters[cluster] += 1
for id in seq_clusters:
cluster = seq_clusters[id]
if not id in refined_clusters:
refined_clusters[id] = dict()
for contig_id in cluster:
if contig_id in circular_contigs and len(cluster) > 1 and (
id in replicon_clusters and replicon_clusters[id] > 1):
if not clust_id in refined_clusters:
refined_clusters["Novel_" + str(clust_id)] = dict()
refined_clusters["Novel_" +
str(clust_id)][contig_id] = cluster[contig_id]
clust_id += 1
continue
refined_clusters[id][contig_id] = cluster[contig_id]
seq_clusters = refined_clusters
m = mash()
mash_distances = dict()
mash_top_dists = dict()
contig_report = list()
results_fh = open(contig_report_file, 'w')
results_fh.write("file_id\tcluster_id\tcontig_id\tcontig_length\tcircularity_status\trep_type\t" \
"rep_type_accession\trelaxase_type\trelaxase_type_accession\tmash_nearest_neighbor\t"
" mash_neighbor_distance\trepetitive_dna_id\tmatch_type\tscore\tcontig_match_start\tcontig_match_end\n")
filter_list = dict()
counter = 0
for cluster in seq_clusters:
clusters = seq_clusters[cluster]
total_cluster_length = 0
count_seqs = len(clusters)
count_rep = 0
count_small = 0
temp = dict()
for contig_id in clusters:
temp[contig_id] = ''
if contig_id in repetitive_contigs:
count_rep += 1
length = len(clusters[contig_id])
total_cluster_length += length
if length < 3000:
count_small += 1
if count_rep == count_seqs or (
float(count_rep) / count_seqs * 100 > 50
and count_small == count_seqs) or total_cluster_length < 1500:
continue
for contig_id in temp:
filter_list[contig_id] = ''
cluster_file = os.path.join(tmp_dir,
'clust_' + str(cluster) + '.fasta')
mash_file = os.path.join(tmp_dir, 'clust_' + str(cluster) + '.txt')
write_fasta_dict(clusters, cluster_file)
mashfile_handle = open(mash_file, 'w')
m.run_mash(mash_db, cluster_file, mashfile_handle)
mash_results = m.read_mash(mash_file)
mash_top_hit = getMashBestHit(mash_results)
# delete low scoring clusters
if float(mash_top_hit['mash_hit_score']) > 0.05:
skip = True
for contig_id in clusters:
if contig_id in replicon_contigs:
skip = False
break
if contig_id in circular_contigs:
skip = False
break
if contig_id in mob_contigs:
skip = False
break
if skip:
for contig_id in clusters:
del (filter_list[contig_id])
continue
new_clust_file = None
if os.path.isfile(cluster_file):
if float(mash_top_hit['mash_hit_score']) < 0.05:
cluster = mash_top_hit['clustid']
new_clust_file = os.path.join(out_dir,
'plasmid_' + cluster + ".fasta")
else:
cluster = 'novel_' + str(counter)
new_clust_file = os.path.join(out_dir,
'plasmid_' + cluster + ".fasta")
counter += 1
if os.path.isfile(new_clust_file):
temp_fh = open(cluster_file, 'r')
data = temp_fh.read()
temp_fh.close()
temp_fh = open(new_clust_file, 'a')
temp_fh.write(data)
temp_fh.close()
mash_file = os.path.join(tmp_dir,
'clust_' + str(cluster) + '.txt')
mashfile_handle = open(mash_file, 'w')
m.run_mash(mash_db, cluster_file, mashfile_handle)
mash_results = m.read_mash(mash_file)
mash_top_hit = getMashBestHit(mash_results)
else:
os.rename(cluster_file, new_clust_file)
if new_clust_file is not None:
plasmid_files[new_clust_file] = ''
for contig_id in clusters:
found_replicon_string = ''
found_replicon_id_string = ''
found_mob_string = ''
found_mob_id_string = ''
contig_status = 'Incomplete'
if contig_id in circular_contigs:
contig_status = 'Circular'
if contig_id in replicon_contigs:
rep_ids = dict()
rep_hit_ids = dict()
for hit_id in replicon_contigs[contig_id]:
id, rep_type = hit_id.split('|')
rep_ids[rep_type] = ''
rep_hit_ids[id] = ''
found_replicon_string = ','.join(list(rep_ids.keys()))
found_replicon_id_string = ','.join(list(rep_hit_ids.keys()))
if contig_id in mob_contigs:
mob_ids = dict()
mob_hit_ids = dict()
for hit_id in mob_contigs[contig_id]:
id, mob_type = hit_id.split('|')
mob_ids[mob_type] = ''
mob_hit_ids[id] = ''
found_mob_string = ','.join(list(mob_ids.keys()))
found_mob_id_string = ','.join(list(mob_hit_ids.keys()))
rep_dna_info = "\t\t\t\t"
if contig_id in repetitive_dna:
rep_dna_info = repetitive_dna[contig_id]
results_fh.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".format(
file_id, cluster, contig_id, len(clusters[contig_id]),
contig_status, found_replicon_string,
found_replicon_id_string, found_mob_string,
found_mob_id_string, mash_top_hit['top_hit'],
mash_top_hit['mash_hit_score'], rep_dna_info))
chr_contigs = dict()
for contig_id in contig_seqs:
if contig_id not in filter_list:
chr_contigs[contig_id] = contig_seqs[contig_id]
rep_dna_info = "\t\t\t\t"
if contig_id in repetitive_dna:
rep_dna_info = repetitive_dna[contig_id]
contig_status = 'Incomplete'
if contig_id in circular_contigs:
contig_status = 'Circular'
results_fh.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".format(
file_id, 'chromosome', contig_id,
len(contig_seqs[contig_id]), contig_status, '', '', '', '',
'', '', rep_dna_info))
results_fh.close()
write_fasta_dict(chr_contigs, chromosome_file)
if args.run_typer:
mobtyper_results = "file_id\tnum_contigs\ttotal_length\tgc\t" \
"rep_type(s)\trep_type_accession(s)\t" \
"relaxase_type(s)\trelaxase_type_accession(s)\t" \
"mpf_type\tmpf_type_accession(s)\t" \
"orit_type(s)\torit_accession(s)\tPredictedMobility\t" \
"mash_nearest_neighbor\tmash_neighbor_distance\tmash_neighbor_cluster\n"
for file in plasmid_files:
mobtyper_results = mobtyper_results + "{}".format(
run_mob_typer(
file, out_dir, str(num_threads),
database_dir=database_dir))
fh = open(mobtyper_results_file, 'w')
fh.write(mobtyper_results)
fh.close()
if not keep_tmp:
shutil.rmtree(tmp_dir)
def main():
args = parse_args()
if args.debug:
logger = init_console_logger(3)
else:
logger = init_console_logger(2)
logger.info('Running Mob-typer version {}'.format(__version__))
logger.info('Processing fasta file {}'.format(args.infile))
if not os.path.isfile(args.infile):
logger.info('Error, fasta file does not exist {}'.format(args.infile))
sys.exit()
if not args.analysis_dir:
tmp_dir = tempfile.TemporaryDirectory(dir=tempfile.gettempdir()).name
else:
tmp_dir = args.analysis_dir
if not os.path.isdir(tmp_dir):
os.mkdir(tmp_dir, 0o755)
if not isinstance(args.num_threads, int):
logger.info(
'Error number of threads must be an integer, you specified "{}"'.
format(args.num_threads))
database_dir = os.path.abspath(args.database_directory)
if args.sample_id is None:
sample_id = re.sub(r"\.(fasta|fa|fas){1,1}", "",
os.path.basename(args.infile))
else:
sample_id = args.sample_id
# Script arguments
input_fasta = args.infile
report_file = args.out_file
num_threads = int(args.num_threads)
keep_tmp = args.keep_tmp
if args.multi:
multi = True
else:
multi = False
if not (args.primary_cluster_dist >= 0 and args.primary_cluster_dist <= 1):
logging.error(
'Error distance thresholds must be between 0 - 1: {}'.format(
args.primary_cluster_dist))
sys.exit()
else:
primary_distance = float(args.primary_cluster_dist)
if not (args.secondary_cluster_dist >= 0
and args.secondary_cluster_dist <= 1):
logging.error(
'Error distance thresholds must be between 0 - 1: {}'.format(
args.secondary_cluster_dist))
sys.exit()
else:
secondary_distance = float(args.secondary_cluster_dist)
if database_dir == default_database_dir:
mob_ref = args.plasmid_mob
mash_db = args.plasmid_mash_db
replicon_ref = args.plasmid_replicons
plasmid_meta = args.plasmid_meta
mpf_ref = args.plasmid_mpf
plasmid_orit = args.plasmid_orit
verify_init(logger, database_dir)
else:
mob_ref = os.path.join(database_dir, 'mob.proteins.faa')
mash_db = os.path.join(database_dir, 'ncbi_plasmid_full_seqs.fas.msh')
replicon_ref = os.path.join(database_dir, 'rep.dna.fas')
plasmid_meta = os.path.join(database_dir, 'clusters.txt')
mpf_ref = os.path.join(database_dir, 'mpf.proteins.faa')
plasmid_orit = os.path.join(database_dir, 'orit.fas')
LIT_PLASMID_TAXONOMY_FILE = os.path.join(
database_dir, "host_range_literature_plasmidDB.txt")
NCBI_PLASMID_TAXONOMY_FILE = plasmid_meta
fixed_fasta = os.path.join(tmp_dir, 'fixed.input.fasta')
replicon_blast_results = os.path.join(tmp_dir,
'replicon_blast_results.txt')
mob_blast_results = os.path.join(tmp_dir, 'mobtyper_blast_results.txt')
mpf_blast_results = os.path.join(tmp_dir, 'mpf_blast_results.txt')
orit_blast_results = os.path.join(tmp_dir, 'orit_blast_results.txt')
repetitive_blast_results = os.path.join(tmp_dir,
'repetitive_blast_results.txt')
if os.path.isfile(mob_blast_results):
os.remove(mob_blast_results)
if os.path.isfile(mpf_blast_results):
os.remove(mpf_blast_results)
if os.path.isfile(orit_blast_results):
os.remove(orit_blast_results)
if os.path.isfile(replicon_blast_results):
os.remove(replicon_blast_results)
# Input numeric params
min_rep_ident = float(args.min_rep_ident)
min_mob_ident = float(args.min_mob_ident)
min_ori_ident = float(args.min_rep_ident)
min_mpf_ident = float(args.min_mob_ident)
idents = {
'min_rep_ident': min_rep_ident,
'min_mob_ident': min_mob_ident,
'min_ori_ident': min_ori_ident
}
for param in idents:
value = float(idents[param])
if value < 60:
logger.error(
"Error: {} is too low, please specify an integer between 70 - 100"
.format(param))
sys.exit(-1)
if value > 100:
logger.error(
"Error: {} is too high, please specify an integer between 70 - 100"
.format(param))
sys.exit(-1)
min_rep_cov = float(args.min_rep_cov)
min_mob_cov = float(args.min_mob_cov)
min_ori_cov = float(args.min_rep_cov)
min_mpf_cov = float(args.min_mob_cov)
covs = {
'min_rep_cov': min_rep_cov,
'min_mob_cov': min_mob_cov,
'min_con_cov': min_ori_cov,
'min_rpp_cov': min_ori_cov
}
for param in covs:
value = float(covs[param])
if value < 60:
logger.error(
"Error: {} is too low, please specify an integer between 50 - 100"
.format(param))
sys.exit(-1)
if value > 100:
logger.error(
"Error: {} is too high, please specify an integer between 50 - 100"
.format(param))
sys.exit(-1)
min_rep_evalue = float(args.min_rep_evalue)
min_mob_evalue = float(args.min_mob_evalue)
min_ori_evalue = float(args.min_rep_evalue)
min_mpf_evalue = float(args.min_mob_evalue)
evalues = {
'min_rep_evalue': min_rep_evalue,
'min_mob_evalue': min_mob_evalue,
'min_con_evalue': min_ori_evalue
}
for param in evalues:
value = float(evalues[param])
if value > 1:
logger.error(
"Error: {} is too high, please specify an float evalue between 0 to 1"
.format(param))
sys.exit(-1)
check_dependencies(logger)
needed_dbs = [replicon_ref, mob_ref, mash_db, mpf_ref]
for db in needed_dbs:
if (not os.path.isfile(db)):
logger.info('Warning! Needed database missing "{}"'.format(db))
mob_suite.mob_init.main()
if not os.path.isdir(tmp_dir):
os.mkdir(tmp_dir, 0o755)
# Test that ETE3 db is ok and lock process check
dbstatus = ETE3_db_status_check(1, ETE3_LOCK_FILE, ETE3DBTAXAFILE, logging)
if dbstatus == False:
logging.error(
"Exiting due to lock file not removed: {}".format(ETE3_LOCK_FILE))
sys.exit(-1)
# Get cluster information
reference_sequence_meta = read_sequence_info(plasmid_meta,
MOB_CLUSTER_INFO_HEADER)
# initilize master record tracking
fix_fasta_header(input_fasta, fixed_fasta)
contig_seqs = read_fasta_dict(fixed_fasta)
contig_info = {}
for id in contig_seqs:
seq = contig_seqs[id]
contig_info[id] = {}
for feature in MOB_TYPER_REPORT_HEADER:
contig_info[id][feature] = ''
contig_info[id]['md5'] = calc_md5(seq)
contig_info[id]['gc'] = GC(seq)
contig_info[id]['size'] = len(seq)
contig_info[id]['contig_id'] = id
contig_info[id]['sample_id'] = sample_id
# Makeblastdb
blast_runner = BlastRunner(fixed_fasta, tmp_dir)
build_success = blast_runner.makeblastdb(fixed_fasta,
'nucl',
logging=logging)
if build_success == False:
logging.error(
"Could not build blast database, check error messages..cannot continue"
)
sys.exit()
# run individual marker blasts
contig_info = identify_biomarkers(contig_info, fixed_fasta, tmp_dir, 25, logging, \
replicon_ref, min_rep_ident, min_rep_cov, min_rep_evalue, replicon_blast_results, \
mob_ref, min_mob_ident, min_mob_cov, min_mob_evalue, mob_blast_results, \
mpf_ref, min_mpf_ident, min_mpf_cov, min_mpf_evalue, mpf_blast_results, \
None, None, None, None, \
plasmid_orit, orit_blast_results, repetitive_blast_results, \
num_threads=1)
m = mash()
mobtyper_results = []
mash_input_fasta = fixed_fasta + '.msh'
ncbi = dict_from_alt_key_list(
read_file_to_dict(NCBI_PLASMID_TAXONOMY_FILE,
MOB_CLUSTER_INFO_HEADER,
separater="\t"), "sample_id")
lit = dict_from_alt_key_list(
read_file_to_dict(LIT_PLASMID_TAXONOMY_FILE,
LIT_PLASMID_TAXONOMY_HEADER,
separater="\t"), "sample_id")
if multi:
m.mashsketch(input_fasta=fixed_fasta,
output_path=mash_input_fasta,
sketch_ind=True,
num_threads=num_threads)
mash_results = parseMash(
m.run_mash(reference_db=mash_db,
input_fasta=mash_input_fasta,
table=False,
num_threads=num_threads))
for seq_id in mash_results:
record = {}
for field in MOB_TYPER_REPORT_HEADER:
if field in contig_info[seq_id]:
record[field] = contig_info[seq_id][field]
else:
record[field] = ''
record['sample_id'] = seq_id
record['num_contigs'] = 1
distances = OrderedDict(
sorted(mash_results[seq_id].items(),
key=itemgetter(1),
reverse=False))
for mash_neighbor_id in distances:
dist = distances[mash_neighbor_id]
if mash_neighbor_id not in reference_sequence_meta:
continue
else:
record['mash_nearest_neighbor'] = mash_neighbor_id
record['mash_neighbor_distance'] = dist
record['primary_cluster_id'] = reference_sequence_meta[
mash_neighbor_id]['primary_cluster_id']
record['secondary_cluster_id'] = reference_sequence_meta[
mash_neighbor_id]['secondary_cluster_id']
record[
'mash_neighbor_identification'] = reference_sequence_meta[
mash_neighbor_id]['organism']
break
mobtyper_results.append(record)
else:
m.mashsketch(input_fasta=fixed_fasta,
output_path=mash_input_fasta,
sketch_ind=False,
num_threads=num_threads)
mash_results = parseMash(
m.run_mash(reference_db=mash_db,
input_fasta=mash_input_fasta,
table=False,
num_threads=num_threads))
record = {}
for field in MOB_TYPER_REPORT_HEADER:
record[field] = ''
record['sample_id'] = sample_id
fastaSeqStats = calcFastaStats(fixed_fasta)
record['md5'] = fastaSeqStats['md5']
record['total_length'] = fastaSeqStats['size']
record['num_contigs'] = fastaSeqStats['num_seq']
record['gc'] = fastaSeqStats['gc_content']
record['mash_nearest_neighbor'] = '-'
record['mash_neighbor_distance'] = 1
record['primary_cluster_id'] = '-'
record['secondary_cluster_id'] = '-'
record['mash_neighbor_identification'] = '-'
for seq_id in mash_results:
distances = OrderedDict(
sorted(mash_results[seq_id].items(),
key=itemgetter(1),
reverse=False))
mash_neighbor_id = next(iter(distances))
dist = distances[mash_neighbor_id]
if mash_neighbor_id not in reference_sequence_meta:
continue
record['mash_nearest_neighbor'] = mash_neighbor_id
record['mash_neighbor_distance'] = dist
record['primary_cluster_id'] = reference_sequence_meta[
mash_neighbor_id]['primary_cluster_id']
record['secondary_cluster_id'] = reference_sequence_meta[
mash_neighbor_id]['secondary_cluster_id']
record['mash_neighbor_identification'] = reference_sequence_meta[
mash_neighbor_id]['organism']
record['rep_type(s)'] = []
record['rep_type_accession(s)'] = []
record['relaxase_type(s)'] = []
record['relaxase_type_accession(s)'] = []
record['mpf_type'] = []
record['mpf_type_accession(s)'] = []
record['orit_type(s)'] = []
record['orit_accession(s)'] = []
for seq_id in contig_info:
record['rep_type(s)'].append(contig_info[seq_id]['rep_type(s)'])
record['rep_type_accession(s)'].append(
contig_info[seq_id]['rep_type_accession(s)'])
record['relaxase_type(s)'].append(
contig_info[seq_id]['relaxase_type(s)'])
record['relaxase_type_accession(s)'].append(
contig_info[seq_id]['relaxase_type_accession(s)'])
record['mpf_type'].append(contig_info[seq_id]['mpf_type'])
record['mpf_type_accession(s)'].append(
contig_info[seq_id]['mpf_type_accession(s)'])
record['orit_type(s)'].append(contig_info[seq_id]['orit_type(s)'])
record['orit_accession(s)'].append(
contig_info[seq_id]['orit_accession(s)'])
for field in record:
tmp = []
if record[field] == None:
continue
if isinstance(record[field], list):
length = len(record[field])
for i in range(0, length):
tmp += record[field][i].split(',')
elif isinstance(record[field], str) and len(record[field]) > 0:
tmp += record[field].split(',')
if len(tmp) > 0:
record[field] = []
for d in tmp:
if len(d) > 0:
record[field].append(d)
mobtyper_results.append(record)
for i in range(0, len(mobtyper_results)):
record = mobtyper_results[i]
bio_markers = sort_biomarkers({
0: {
'types': record['rep_type(s)'],
'acs': record['rep_type_accession(s)']
},
1: {
'types': record['relaxase_type(s)'],
'acs': record['relaxase_type_accession(s)']
},
2: {
'types': record['mpf_type'],
'acs': record['mpf_type_accession(s)']
},
3: {
'types': record['orit_type(s)'],
'acs': record['orit_accession(s)']
},
})
record['rep_type(s)'] = bio_markers[0]['types']
record['rep_type_accession(s)'] = bio_markers[0]['acs']
record['relaxase_type(s)'] = bio_markers[1]['types']
record['relaxase_type_accession(s)'] = bio_markers[1]['acs']
record['mpf_type'] = bio_markers[2]['types']
record['mpf_type_accession(s)'] = bio_markers[2]['acs']
record['orit_type(s)'] = bio_markers[3]['types']
record['orit_accession(s)'] = bio_markers[3]['acs']
if (isinstance(record['mash_neighbor_distance'], float) or isinstance(
record['mash_neighbor_distance'],
int)) and record['mash_neighbor_distance'] <= primary_distance:
mob_cluster_id = record['primary_cluster_id']
else:
mob_cluster_id = None
#Patches that sometimes results are concatonated into strings if contigs are merged into a single results
if isinstance(record['rep_type(s)'], list):
record['rep_type(s)'] = ",".join(record['rep_type(s)'])
if isinstance(record['relaxase_type_accession(s)'], list):
record['relaxase_type_accession(s)'] = ",".join(
record['relaxase_type_accession(s)'])
host_range = hostrange(record['rep_type(s)'].split(','),
record['relaxase_type_accession(s)'].split(','),
mob_cluster_id, ncbi, lit)
for field in host_range:
record[field] = host_range[field]
if isinstance(record['mpf_type'], list):
record['mpf_type'] = determine_mpf_type(record['mpf_type'])
elif isinstance(record['mpf_type'], str):
record['mpf_type'] = determine_mpf_type(
record['mpf_type'].split(','))
for field in record:
if isinstance(record[field], list):
record[field] = ",".join(record[field])
record['predicted_mobility'] = 'non-mobilizable'
if len(record['relaxase_type(s)']) > 0 and len(record['mpf_type']):
record['predicted_mobility'] = 'conjugative'
elif len(record['relaxase_type(s)']) > 0 or len(
record['orit_type(s)']) > 0:
record['predicted_mobility'] = 'mobilizable'
mobtyper_results[i] = record
writeReport(mobtyper_results, MOB_TYPER_REPORT_HEADER, report_file)
if not keep_tmp:
shutil.rmtree(tmp_dir)
logger.info(
"MOB-typer completed and results written to {}".format(report_file))
def main():
args = parse_args()
if args.debug:
init_console_logger(3)
logging.info('Running plasmid detector v. {}'.format(__version__))
if not args.outdir:
logging.info(
'Error, no output directory specified, please specify one')
sys.exit()
if not args.infile:
logging.info('Error, no fasta specified, please specify one')
sys.exit()
logging.info('Processing fasta file {}'.format(args.infile))
logging.info('Analysis directory {}'.format(args.outdir))
if not os.path.isdir(args.outdir):
os.mkdir(args.outdir, 0o755)
verify_init(logging)
plasmid_files = dict()
input_fasta = args.infile
out_dir = args.outdir
num_threads = args.num_threads
tmp_dir = os.path.join(out_dir, '__tmp')
file_id = os.path.basename(input_fasta)
fixed_fasta = os.path.join(tmp_dir, 'fixed.input.fasta')
chromosome_file = os.path.join(out_dir, 'chromosome.fasta')
replicon_blast_results = os.path.join(tmp_dir,
'replicon_blast_results.txt')
mob_blast_results = os.path.join(tmp_dir, 'mobrecon_blast_results.txt')
repetitive_blast_results = os.path.join(tmp_dir,
'repetitive_blast_results.txt')
contig_blast_results = os.path.join(tmp_dir, 'contig_blast_results.txt')
min_ident = args.min_ident
min_cov = args.min_cov
evalue = args.evalue
min_length = args.min_length
plasmid_ref_db = args.plasmid_db
replicon_ref = args.plasmid_replicons
mob_ref = args.plasmid_mob
mash_db = args.plasmid_mash_db
repetitive_mask_file = args.repetitive_mask
contig_report_file = os.path.join(out_dir, 'contig_report.txt')
minimus_prefix = os.path.join(tmp_dir, 'minimus')
filtered_blast = os.path.join(tmp_dir, 'filtered_blast.txt')
repetitive_blast_report = os.path.join(out_dir,
'repetitive_blast_report.txt')
mobtyper_results_file = os.path.join(out_dir,
'mobtyper_aggregate_report.txt')
keep_tmp = args.keep_tmp
if args.run_circlator == 'False' or args.run_circlator == 'false':
run_circlator = False
else:
run_circlator = True
if args.unicycler_contigs == 'False' or args.unicycler_contigs == 'false':
unicycler_contigs = False
else:
unicycler_contigs = True
if not isinstance(args.num_threads, int):
logging.info(
'Error number of threads must be an integer, you specified "{}"'.
format(args.num_threads))
logging.info('Creating tmp working directory {}'.format(tmp_dir))
if not os.path.isdir(tmp_dir):
os.mkdir(tmp_dir, 0o755)
logging.info(
'Writing cleaned header input fasta file from {} to {}'.format(
input_fasta, fixed_fasta))
fix_fasta_header(input_fasta, fixed_fasta)
contig_seqs = read_fasta_dict(fixed_fasta)
logging.info('Running replicon blast on {}'.format(replicon_ref))
replicon_contigs = getRepliconContigs(
replicon_blast(replicon_ref,
fixed_fasta,
80,
80,
evalue,
tmp_dir,
replicon_blast_results,
num_threads=num_threads))
logging.info('Running relaxase blast on {}'.format(mob_ref))
mob_contigs = getRepliconContigs(
mob_blast(mob_ref,
fixed_fasta,
80,
80,
evalue,
tmp_dir,
mob_blast_results,
num_threads=num_threads))
logging.info('Running contig blast on {}'.format(plasmid_ref_db))
contig_blast(fixed_fasta, plasmid_ref_db, args.min_ident, args.min_cov,
evalue, min_length, tmp_dir, contig_blast_results)
pcl_clusters = contig_blast_group(filtered_blast, 10)
logging.info(
'Running repetitive contig masking blast on {}'.format(mob_ref))
repetitive_contigs = repetitive_blast(fixed_fasta,
repetitive_mask_file,
min_ident,
min_cov,
evalue,
300,
tmp_dir,
repetitive_blast_results,
num_threads=num_threads)
circular_contigs = dict()
logging.info('Running circlator minimus2 on {}'.format(fixed_fasta))
if run_circlator:
circular_contigs = circularize(fixed_fasta, minimus_prefix)
if unicycler_contigs:
for seqid in contig_seqs:
if 'circular=true' in seqid:
circular_contigs[seqid] = ''
repetitive_dna = dict()
results_fh = open(repetitive_blast_report, 'w')
results_fh.write(
"contig_id\tmatch_id\tmatch_type\tscore\tcontig_match_start\tcontig_match_end\n"
)
for contig_id in repetitive_contigs:
match_info = repetitive_contigs[contig_id]['id'].split('|')
repetitive_dna[contig_id] = "{}\t{}\t{}\t{}\t{}".format(
match_info[1], match_info[len(match_info) - 1],
repetitive_contigs[contig_id]['score'],
repetitive_contigs[contig_id]['contig_start'],
repetitive_contigs[contig_id]['contig_end'])
results_fh.write("{}\t{}\t{}\t{}\t{}\t{}\n".format(
contig_id, match_info[1], match_info[len(match_info) - 1],
repetitive_contigs[contig_id]['score'],
repetitive_contigs[contig_id]['contig_start'],
repetitive_contigs[contig_id]['contig_end']))
results_fh.close()
seq_clusters = dict()
cluster_bitscores = dict()
for seqid in pcl_clusters:
cluster_id = list(pcl_clusters[seqid].keys())[0]
bitscore = pcl_clusters[seqid][cluster_id]
cluster_bitscores[cluster_id] = bitscore
sorted_cluster_bitscores = sorted(list(cluster_bitscores.items()),
key=operator.itemgetter(1))
sorted_cluster_bitscores.reverse()
contigs_assigned = dict()
for cluster_id, bitscore in sorted_cluster_bitscores:
if not cluster_id in seq_clusters:
seq_clusters[cluster_id] = dict()
for seqid in pcl_clusters:
if not cluster_id in pcl_clusters[seqid]:
continue
if seqid in contig_seqs and seqid not in contigs_assigned:
seq_clusters[cluster_id][seqid] = contig_seqs[seqid]
contigs_assigned[seqid] = cluster_id
# Add sequences with known replicons regardless of whether they belong to a mcl cluster
clust_id = 0
refined_clusters = dict()
for contig_id in mob_contigs:
if not contig_id in pcl_clusters:
if contig_id in contig_seqs:
if not clust_id in seq_clusters:
seq_clusters["Novel_" + str(clust_id)] = dict()
if not contig_id in pcl_clusters:
pcl_clusters[contig_id] = dict()
pcl_clusters[contig_id]["Novel_" + str(clust_id)] = 0
seq_clusters["Novel_" +
str(clust_id)][contig_id] = contig_seqs[contig_id]
clust_id += 1
# Add sequences with known relaxases regardless of whether they belong to a mcl cluster
count_replicons = dict()
for contig_id in replicon_contigs:
if not contig_id in pcl_clusters:
if contig_id in contig_seqs:
if not clust_id in seq_clusters:
seq_clusters["Novel_" + str(clust_id)] = dict()
if not contig_id in pcl_clusters:
pcl_clusters[contig_id] = dict()
pcl_clusters[contig_id]["Novel_" + str(clust_id)] = dict()
seq_clusters["Novel_" +
str(clust_id)][contig_id] = contig_seqs[contig_id]
clust_id += 1
refined_clusters = dict()
# split out circular sequences from each other
replicon_clusters = dict()
for contig_id in replicon_contigs:
for hit_id in replicon_contigs[contig_id]:
id, rep_type = hit_id.split('|')
cluster = list(pcl_clusters[contig_id].keys())[0]
if not cluster in replicon_clusters:
replicon_clusters[cluster] = 0
replicon_clusters[cluster] += 1
for id in seq_clusters:
cluster = seq_clusters[id]
if not id in refined_clusters:
refined_clusters[id] = dict()
for contig_id in cluster:
if contig_id in circular_contigs and len(cluster) > 1 and (
id in replicon_clusters and replicon_clusters[id] > 1):
if not clust_id in refined_clusters:
refined_clusters["Novel_" + str(clust_id)] = dict()
refined_clusters["Novel_" +
str(clust_id)][contig_id] = cluster[contig_id]
clust_id += 1
continue
refined_clusters[id][contig_id] = cluster[contig_id]
seq_clusters = refined_clusters
m = mash()
mash_distances = dict()
mash_top_dists = dict()
contig_report = list()
results_fh = open(contig_report_file, 'w')
results_fh.write("file_id\tcluster_id\tcontig_id\tcontig_length\tcircularity_status\trep_type\t" \
"rep_type_accession\trelaxase_type\trelaxase_type_accession\tmash_nearest_neighbor\t"
" mash_neighbor_distance\trepetitive_dna_id\tmatch_type\tscore\tcontig_match_start\tcontig_match_end\n")
filter_list = dict()
counter = 0
for cluster in seq_clusters:
clusters = seq_clusters[cluster]
total_cluster_length = 0
count_seqs = len(clusters)
count_rep = 0
count_small = 0
temp = dict()
for contig_id in clusters:
temp[contig_id] = ''
if contig_id in repetitive_contigs:
count_rep += 1
length = len(clusters[contig_id])
total_cluster_length += length
if length < 3000:
count_small += 1
if count_rep == count_seqs or (
float(count_rep) / count_seqs * 100 > 50
and count_small == count_seqs) or total_cluster_length < 1500:
continue
for contig_id in temp:
filter_list[contig_id] = ''
cluster_file = os.path.join(tmp_dir,
'clust_' + str(cluster) + '.fasta')
mash_file = os.path.join(tmp_dir, 'clust_' + str(cluster) + '.txt')
write_fasta_dict(clusters, cluster_file)
mashfile_handle = open(mash_file, 'w')
m.run_mash(mash_db, cluster_file, mashfile_handle)
mash_results = m.read_mash(mash_file)
mash_top_hit = getMashBestHit(mash_results)
# delete low scoring clusters
if float(mash_top_hit['mash_hit_score']) > 0.05:
skip = True
for contig_id in clusters:
if contig_id in replicon_contigs:
skip = False
break
if contig_id in circular_contigs:
skip = False
break
if contig_id in mob_contigs:
skip = False
break
if skip:
for contig_id in clusters:
del (filter_list[contig_id])
continue
new_clust_file = None
if os.path.isfile(cluster_file):
if float(mash_top_hit['mash_hit_score']) < 0.05:
cluster = mash_top_hit['clustid']
new_clust_file = os.path.join(out_dir,
'plasmid_' + cluster + ".fasta")
else:
cluster = 'novel_' + str(counter)
new_clust_file = os.path.join(out_dir,
'plasmid_' + cluster + ".fasta")
counter += 1
if os.path.isfile(new_clust_file):
temp_fh = open(cluster_file, 'r')
data = temp_fh.read()
temp_fh.close()
temp_fh = open(new_clust_file, 'a')
temp_fh.write(data)
temp_fh.close()
mash_file = os.path.join(tmp_dir,
'clust_' + str(cluster) + '.txt')
mashfile_handle = open(mash_file, 'w')
m.run_mash(mash_db, cluster_file, mashfile_handle)
mash_results = m.read_mash(mash_file)
mash_top_hit = getMashBestHit(mash_results)
else:
os.rename(cluster_file, new_clust_file)
if new_clust_file is not None:
plasmid_files[new_clust_file] = ''
for contig_id in clusters:
found_replicon_string = ''
found_replicon_id_string = ''
found_mob_string = ''
found_mob_id_string = ''
contig_status = 'Incomplete'
if contig_id in circular_contigs:
contig_status = 'Circular'
if contig_id in replicon_contigs:
rep_ids = dict()
rep_hit_ids = dict()
for hit_id in replicon_contigs[contig_id]:
id, rep_type = hit_id.split('|')
rep_ids[rep_type] = ''
rep_hit_ids[id] = ''
found_replicon_string = ','.join(list(rep_ids.keys()))
found_replicon_id_string = ','.join(list(rep_hit_ids.keys()))
if contig_id in mob_contigs:
mob_ids = dict()
mob_hit_ids = dict()
for hit_id in mob_contigs[contig_id]:
id, mob_type = hit_id.split('|')
mob_ids[mob_type] = ''
mob_hit_ids[id] = ''
found_mob_string = ','.join(list(mob_ids.keys()))
found_mob_id_string = ','.join(list(mob_hit_ids.keys()))
rep_dna_info = "\t\t\t\t"
if contig_id in repetitive_dna:
rep_dna_info = repetitive_dna[contig_id]
results_fh.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".format(
file_id, cluster, contig_id, len(clusters[contig_id]),
contig_status, found_replicon_string,
found_replicon_id_string, found_mob_string,
found_mob_id_string, mash_top_hit['top_hit'],
mash_top_hit['mash_hit_score'], rep_dna_info))
chr_contigs = dict()
for contig_id in contig_seqs:
if contig_id not in filter_list:
chr_contigs[contig_id] = contig_seqs[contig_id]
rep_dna_info = "\t\t\t\t"
if contig_id in repetitive_dna:
rep_dna_info = repetitive_dna[contig_id]
contig_status = 'Incomplete'
if contig_id in circular_contigs:
contig_status = 'Circular'
results_fh.write(
"{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n".format(
file_id, 'chromosome', contig_id,
len(contig_seqs[contig_id]), contig_status, '', '', '', '',
'', '', rep_dna_info))
results_fh.close()
write_fasta_dict(chr_contigs, chromosome_file)
if args.run_typer:
mobtyper_results = "file_id\tnum_contigs\ttotal_length\tgc\t" \
"rep_type(s)\trep_type_accession(s)\t" \
"relaxase_type(s)\trelaxase_type_accession(s)\t" \
"mpf_type\tmpf_type_accession(s)\t" \
"orit_type(s)\torit_accession(s)\tPredictedMobility\t" \
"mash_nearest_neighbor\tmash_neighbor_distance\tmash_neighbor_cluster\n"
for file in plasmid_files:
mobtyper_results = mobtyper_results + "{}".format(
run_mob_typer(file, out_dir, str(num_threads)))
fh = open(mobtyper_results_file, 'w')
fh.write(mobtyper_results)
fh.close()
if not keep_tmp:
shutil.rmtree(tmp_dir)
def main():
args = parse_args()
if args.debug:
logging = init_console_logger(3)
else:
logging = init_console_logger(2)
logging.info(
'Running Mob-Suite Clustering toolkit v. {}'.format(__version__))
logging.info('Processing fasta file {}'.format(args.infile))
logging.info('Analysis directory {}'.format(args.outdir))
check_dependencies(logging)
input_fasta = args.infile
if not os.path.isfile(input_fasta):
logging.error('Error, input fasta specified does not exist: {}'.format(
input_fasta))
sys.exit()
mob_typer_report_file = args.mob_typer_file
if not os.path.isfile(mob_typer_report_file):
logging.error(
'Error, input metadata file specified does not exist: {}'.format(
mob_typer_report_file))
sys.exit()
mode = str(args.mode).lower()
if mode not in ('update', 'build'):
logging.error(
'Error you have not entered a valid mode of build or update, you entered: {}'
.format(mode))
sys.exit()
out_dir = args.outdir
num_threads = args.num_threads
if not (args.primary_cluster_dist >= 0 and args.primary_cluster_dist <= 1):
logging.error(
'Error distance thresholds must be between 0 - 1: {}'.format(
args.primary_cluster_dist))
sys.exit()
else:
primary_distance = args.primary_cluster_dist
if not (args.secondary_cluster_dist >= 0
and args.secondary_cluster_dist <= 1):
logging.error(
'Error distance thresholds must be between 0 - 1: {}'.format(
args.secondary_cluster_dist))
sys.exit()
else:
secondary_distance = args.secondary_cluster_dist
if not os.path.isdir(out_dir):
logging.info('Creating directory {}'.format(args.outdir))
os.mkdir(out_dir, 0o755)
tmp_dir = os.path.join(out_dir, '__tmp')
if not os.path.isdir(tmp_dir):
logging.info('Creating directory {}'.format(args.outdir))
os.mkdir(tmp_dir, 0o755)
taxonomy_file = args.taxonomy
records = read_file_to_dict(mob_typer_report_file,
MOB_TYPER_REPORT_HEADER,
separater="\t")
seq_ids = []
new_seq_info = {}
duplicate_keys = []
for record in records:
seq_ids.append(record['sample_id'])
if not record['sample_id'] in new_seq_info:
new_seq_info[record['sample_id']] = record
else:
duplicate_keys.append(record['sample_id'])
if len(duplicate_keys) > 0:
logging.error(
"Duplicate sequence identifiers in fasta file. Please make every sequence id unique in the input file before using this tool"
)
logging.error("Duplicate sequence ids: {}".format(
",".join(duplicate_keys)))
sys.exit()
record_identifications = read_file_to_dict(taxonomy_file,
['sample_id', 'organism'],
separater="\t")
organisms = []
for record in record_identifications:
organism = record['organism']
if organism == 'unknown' or organism == '' or organism == 'Unknown':
organism = 'Bacteria'
organisms.append(organism)
seq_id = record['sample_id']
if seq_id in new_seq_info:
new_seq_info[seq_id]['organism'] = organism
taxids = NamesToTaxIDs(organisms)
del (organisms)
for seq_id in new_seq_info:
organism = new_seq_info[seq_id]['organism']
if organism in taxids:
new_seq_info[seq_id]['taxid'] = taxids[organism][0]
else:
new_seq_info[seq_id]['taxid'] = 2
if len(new_seq_info) == 0:
logging.error(
'Error no MOB-typer results for sequences. Sequences must be typed with MOB-typer first'
)
sys.exit()
fasta_dict = read_fasta_dict(input_fasta)
if len(fasta_dict) == 0:
logging.error(
'Error no sequences found in input fasta: {}..cannot continue'.
format(input_fasta))
sys.exit()
key_set_1 = set(seq_ids)
key_set_2 = set(list(fasta_dict.keys()))
if len(list(key_set_1 ^ key_set_2)) > 0:
logging.error(
'Error MOB-typer results: {} and input fasta: {} do not have the same set of identifiers, these must match in order to proceed'
.format(mob_typer_report_file, input_fasta))
logging.error(
'Keys present in MOB-typer results: {} and not in input fasta: {} are: {}'
.format(mob_typer_report_file, input_fasta,
list(key_set_1 - key_set_2)))
logging.error(
'Keys present in MOB-typer results: {} and not in input fasta: {} are: {}'
.format(mob_typer_report_file, input_fasta,
list(key_set_2 - key_set_1)))
sys.exit()
tmp_cluster_file = os.path.join(out_dir, 'clusters.txt')
tmp_ref_fasta_file = os.path.join(tmp_dir, 'references_tmp.fasta')
update_fasta = os.path.join(out_dir, 'references_updated.fasta')
# Sketch and calculate distances within update sequences
if len(fasta_dict) > 1:
mashObj = mash()
mashObj.mashsketch(input_fasta,
input_fasta + ".msh",
num_threads=num_threads)
distance_matrix_file = os.path.join(tmp_dir, 'mash_dist_matrix.txt')
mashfile_handle = open(distance_matrix_file, 'w', encoding="utf-8")
mashfile_handle.write(
mashObj.run_mash(input_fasta + '.msh',
input_fasta + '.msh',
table=True,
num_threads=num_threads).decode())
mashfile_handle.close()
clust_assignments = build_cluster_db(
distance_matrix_file, (primary_distance, secondary_distance))
else:
seq_id = next(iter(fasta_dict))
clust_assignments = {seq_id: [0, 1]}
logging.info('Running MOB-cluster in {} mode'.format(mode))
if mode == 'update':
if args.ref_cluster_file is None:
logging.error(
'Reference fasta file must be specified, please check help for parameter reference'
)
sys.exit()
ref_fasta = args.ref_fasta_file
if not os.path.isfile(ref_fasta):
logging.error(
'Reference fasta file specified does not exist: {}'.format(
ref_fasta))
sys.exit()
if args.ref_cluster_file is None:
logging.error(
'Reference cluster file must be specified, please check help for parameter reference'
)
sys.exit()
ref_cluster_file = args.ref_cluster_file
if not os.path.isfile(ref_cluster_file):
logging.error(
'Reference cluster file specified does not exist: {}'.format(
ref_cluster_file))
sys.exit()
mob_cluster_seq_info = read_sequence_info(ref_cluster_file,
MOB_CLUSTER_INFO_HEADER)
logging.info(
'Running mob-cluster in update mode with input file: {}'.format(
input_fasta))
logging.info(
'Running mob-cluster in update mode with output directory: {}'.
format(out_dir))
logging.info(
'Running mob-cluster in update mode on reference fasta file: {}'.
format(ref_fasta))
logging.info(
'Reading previous cluster reference assignments from : {}'.format(
ref_cluster_file))
shutil.copy(ref_cluster_file, tmp_cluster_file)
shutil.copy(ref_fasta, tmp_ref_fasta_file)
logging.info('Creating new cluster assignments')
new_seq_info = update_existing_db(new_seq_info, mob_cluster_seq_info,
clust_assignments, primary_distance,
secondary_distance, num_threads)
cluster_assignments = {**mob_cluster_seq_info, **new_seq_info}
logging.info(
'Writting cluster assignments to : {}'.format(tmp_cluster_file))
writeClusterAssignments(tmp_cluster_file, MOB_CLUSTER_INFO_HEADER,
cluster_assignments)
shutil.copy(tmp_ref_fasta_file, os.path.join(out_dir, update_fasta))
else:
cluster_acs = convert_num_to_acs(clust_assignments)
for id in cluster_acs:
primary_key = cluster_acs[id][0]
secondary_key = cluster_acs[id][1]
new_seq_info[id]['primary_cluster_id'] = primary_key
new_seq_info[id]['primary_dist'] = primary_distance
new_seq_info[id]['secondary_cluster_id'] = secondary_key
new_seq_info[id]['secondary_dist'] = secondary_distance
writeClusterAssignments(tmp_cluster_file, MOB_CLUSTER_INFO_HEADER,
new_seq_info)
shutil.copy(input_fasta, update_fasta)
logging.info("Sketching new fasta {}".format(update_fasta))
mash_db_file = "{}.msh".format(update_fasta)
mObj = mash()
mObj.mashsketch(update_fasta, mash_db_file, num_threads=num_threads)
logging.info("Building blastdb {}".format(update_fasta))
blast_runner = BlastRunner(update_fasta, '')
blast_runner.makeblastdb(update_fasta, 'nucl', logging=logging)
logging.info("Removing temporary directory")
shutil.rmtree(tmp_dir)
logging.info(
"MOB-cluster completed, analysis results written to {}".format(
out_dir))
