def iterate_handle_threaded(handle, ref, database='data/gsaid.db'):
"""
Function to iterate through handle passed by Chunky bot
:param cursor: sqlite database handler
:param handle: list containing tuples (header, raw seq)
:param ref: file containing reference sequence, default-> NC_04552.fa
"""
_, refseq = gotoh2.convert_fasta(open(ref))[0]
in_queue = queue.Queue()
out_queue = queue.Queue()
# stream records into queue
for h, s in handle:
in_queue.put((h, s, database, refseq))
# pairwise align sequences in queue
for seq in range(0, in_queue.qsize()):
AlignerThread(in_queue, out_queue).start()
# block until queue is processed
in_queue.join()
# update database with aligned sequences
cursor, conn = open_connection(database)
while not out_queue.empty():
header, seq, alignedseq = out_queue.get()
insert_seq(cursor, seq, header, alignedseq)
conn.commit()
out_queue.task_done()
conn.close()
out_queue.join()
def iterate_fasta(fasta, ref, database = 'data/gsaid.db'): """ Function to iterate through fasta file :params: :cursor: sqlite database handler :fasta: file containing sequences :ref: file containing reference sequence, default-> NC_04552.fa """ handle = gotoh2.iter_fasta(open(fasta)) _, refseq = gotoh2.convert_fasta(open(ref))[0] in_queue= queue.Queue() out_queue= queue.Queue() for h, s in handle: in_queue.put((h, s, database, refseq)) for seq in range(0,in_queue.qsize()): alignerThread(in_queue, out_queue).start() print(threading.active_count()) in_queue.join() cursor, conn = open_connection(database) while not out_queue.empty(): header, seq, alignedseq = out_queue.get() insert_seq(cursor, seq, header, alignedseq) conn.commit() out_queue.task_done() conn.close() out_queue.join()
def iterate_fasta(fasta, ref, database='data/gsaid.db'):
"""
Function to iterate through fasta file *non threaded*
:param cursor: sqlite database handler
:param fasta: file containing sequences
:param ref: file containing reference sequence, default-> NC_04552.fa
"""
handle = gotoh2.iter_fasta(open(fasta))
_, refseq = gotoh2.convert_fasta(open(ref))[0]
cursor, conn = open_connection(database)
# stream records into queue
for header, seq in handle:
alignedseq = find_seq(conn, seq, refseq)
insert_seq(cursor, seq, header, alignedseq)
conn.commit()
conn.close()
def iterate_handle(handle, ref, database='data/gsaid.db'):
"""
Function to iterate through list [(header,seq)....] passed by ChunkyBot ##NON THREADED##
:param cursor: sqlite database handler
:param handle: list containing tuples (header, raw seq)
:param ref: file containing reference sequence, default-> NC_04552.fa
"""
_, refseq = gotoh2.convert_fasta(open(ref))[0]
cursor, conn = open_connection(database)
#align and insert into database
for header, seq in handle:
alignedseq = find_seq(conn, seq, refseq)
#if alignedseq returns 0; raw seq length <5,000
if alignedseq == 0:
print('Sequence {} with a length of {} cannot be aligned.'.format(
header, len(seq)))
else:
print('Aligning {}.'.format(header))
insert_seq(cursor, seq, header, alignedseq)
conn.commit()
maxScore = 0
maxIndex = -1
# get the max probability score and its assosciated index
for i in range(len(predictions[index])):
if predictions[index][i] > maxScore:
maxScore = predictions[index][i]
maxIndex = i
score = maxScore
prediction = loaded_model.classes_[maxIndex]
accession = idList[index].split('|')[1]
payload = [accession, prediction, score, pangoLEARN.__version__, 'passed_qc', '']
insert_lineage(cursor, payload)
conn.commit()
if __name__ == '__main__':
args = parse_args()
sequence_handle = gotoh2.convert_fasta(open(args.sequences_file, 'r'))
filtered_handle = filter_seqs(sequence_handle, args.filterout, max_prop_n=0.05, minlen=29000)
if args.header-file == None and args.model-file == None:
classify_and_insert(args.pangolindir+ 'decisionTreeHeaders_v1.joblib', args.pangolindir+'decisionTree_v1.joblib', filtered_handle, args.indicies, args.db)
else:
classify_and_insert(args.header_file, args.model_file, filtered_handle, args.indicies, args.db)