def do_run(args, data_path, match_score, vfs_list, cons_run):
"""
Perform a SeqFindr run
"""
matrix, y_label, exist_ord = [], [], []
in_files = util.get_fasta_files(data_path)
# Reorder if requested
if args.index_file is not None:
in_files = util.order_inputs(args.index_file, in_files)
# Handle and existing run
if args.existing_data is not None:
cleaned, blastxml_tmp = [], []
blast_xml = glob.glob(
os.path.abspath(args.existing_data) + "/BLAST_results/*")
if not cons_run:
for e in blast_xml:
if e.find("cons_DB=") == -1:
blastxml_tmp.append(e)
else:
for e in blast_xml:
if e.find("cons_DB=") != -1:
blastxml_tmp.append(e)
blast_xml = []
blast_xml = blastxml_tmp
for e in blast_xml:
sid = e.split("ID=")[-1].split("_blast.xml")[0]
cleaned.append(sid)
for i in in_files:
for j in cleaned:
if i.find(j) != -1:
y_label.append(j)
break
for i in blast_xml:
for j in y_label:
if i.find(j) != -1:
exist_ord.append(i)
break
in_files = exist_ord
# Make sure XML are right order
for idx, subject in enumerate(in_files):
if args.existing_data is None:
strain_id = blast.make_BLAST_database(subject)
y_label.append(strain_id)
database = os.path.basename(subject)
blast_xml = blast.run_BLAST(
args.seqs_of_interest,
os.path.join(os.getcwd(), "DBs/" + database), args, cons_run)
accepted_hits = blast.parse_BLAST(blast_xml, float(args.tol),
args.careful)
else:
strain_id = y_label[idx]
accepted_hits = blast.parse_BLAST(in_files[idx], float(args.tol),
args.careful)
row = build_matrix_row(vfs_list, accepted_hits, match_score)
row.insert(0, strain_id)
matrix.append(row)
return matrix, y_label
def do_run(args, data_path, match_score, vfs_list, cons_run):
"""
Perform a SeqFindr run
"""
matrix, y_label, exist_ord = [], [], []
in_files = util.get_fasta_files(data_path)
# Reorder if requested
if args.index_file is not None:
in_files = util.order_inputs(args.index_file, in_files)
# Handle and existing run
if args.existing_data is not None:
cleaned, blastxml_tmp = [], []
blast_xml = glob.glob(os.path.abspath(args.existing_data)+"/BLAST_results/*")
if not cons_run:
for e in blast_xml:
if e.find("cons_DB=") == -1:
blastxml_tmp.append(e)
else:
for e in blast_xml:
if e.find("cons_DB=") != -1:
blastxml_tmp.append(e)
blast_xml = []
blast_xml = blastxml_tmp
for e in blast_xml:
sid = e.split("ID=")[-1].split("_blast.xml")[0]
cleaned.append(sid)
for i in in_files:
for j in cleaned:
if i.find(j) != -1:
y_label.append(j)
break
for i in blast_xml:
for j in y_label:
if i.find(j) != -1:
exist_ord.append(i)
break
in_files = exist_ord
# Make sure XML are right order
for idx, subject in enumerate(in_files):
if args.existing_data is None:
strain_id = blast.make_BLAST_database(subject)
y_label.append(strain_id)
database = os.path.basename(subject)
blast_xml = blast.run_BLAST(args.seqs_of_interest, os.path.join(os.getcwd(), "DBs/"+database), args, cons_run)
accepted_hits = blast.parse_BLAST(blast_xml, float(args.tol), args.careful)
else:
strain_id = y_label[idx]
accepted_hits = blast.parse_BLAST(in_files[idx], float(args.tol), args.careful)
row = build_matrix_row(vfs_list, accepted_hits, match_score)
row.insert(0, strain_id)
matrix.append(row)
return matrix, y_label
def strip_bases(args):
"""
Strip the 1st and last 'N' bases from mapping consensuses
Uses:
* args.cons
* args.seqs_of_interest
* arg.strip
To avoid the effects of lead in and lead out coverage resulting in
uncalled bases
:param args: the argparse args containing args.strip value
:type args: argparse args
:rtype: the updated args to reflect the args.cons &
args.seqs_of_interest location
"""
# Get in the fasta files in the consensus directory
fasta_in = util.get_fasta_files(args.cons)
# Build a stripped directory
new_cons_dir = os.path.join(args.cons, 'stripped')
try:
os.mkdir(new_cons_dir)
except OSError:
sys.stderr.write("A stripped directory exists. Overwriting\n")
# Update the args.cons to the stripped directory
args.cons = new_cons_dir
args.strip = int(args.strip)
# Strip the start and end
for fa in fasta_in:
tmp = os.path.basename(fa)
out = os.path.join(args.cons, tmp)
with open(fa, "rU") as fin, open(out, 'w') as fout:
records = SeqIO.parse(fin, "fasta")
for rec in records:
rec.seq = rec.seq[args.strip:-args.strip]
SeqIO.write(rec, fout, "fasta")
# Trim the db as well
tmp = args.seqs_of_interest.split('.')
stripdb = '.'.join(tmp[:-1])+'_trimmed.'+tmp[-1]
with open(args.seqs_of_interest, "rU") as fin, open(stripdb, 'w') as fout:
records = SeqIO.parse(fin, "fasta")
for rec in records:
rec.seq = rec.seq[args.strip:-args.strip]
SeqIO.write(rec, fout, "fasta")
# Update the args.seqs_of_interest
args.seqs_of_interest = stripdb
return args
def strip_bases(args):
"""
Strip the 1st and last 'N' bases from mapping consensuses
Uses:
* args.cons
* args.seqs_of_interest
* arg.strip
To avoid the effects of lead in and lead out coverage resulting in
uncalled bases
:param args: the argparse args containing args.strip value
:type args: argparse args
:rtype: the updated args to reflect the args.cons &
args.seqs_of_interest location
"""
# Get in the fasta files in the consensus directory
fasta_in = util.get_fasta_files(args.cons)
# Build a stripped directory
new_cons_dir = os.path.join(args.cons, 'stripped')
try:
os.mkdir(new_cons_dir)
except OSError:
sys.stderr.write("A stripped directory exists. Overwriting\n")
# Update the args.cons to the stripped directory
args.cons = new_cons_dir
args.strip = int(args.strip)
# Strip the start and end
for fa in fasta_in:
tmp = os.path.basename(fa)
out = os.path.join(args.cons, tmp)
with open(fa, "rU") as fin, open(out, 'w') as fout:
records = SeqIO.parse(fin, "fasta")
for rec in records:
rec.seq = rec.seq[args.strip:-args.strip]
SeqIO.write(rec, fout, "fasta")
# Trim the db as well
tmp = args.seqs_of_interest.split('.')
stripdb = '.'.join(tmp[:-1]) + '_trimmed.' + tmp[-1]
with open(args.seqs_of_interest, "rU") as fin, open(stripdb, 'w') as fout:
records = SeqIO.parse(fin, "fasta")
for rec in records:
rec.seq = rec.seq[args.strip:-args.strip]
SeqIO.write(rec, fout, "fasta")
# Update the args.seqs_of_interest
args.seqs_of_interest = stripdb
return args