import argparse
from collections import Counter
from utils.ClusterIO import sortby, parse, Cluster
from Bio import SeqIO
# Setup parser
parser = argparse.ArgumentParser(description='Procedure to get the genuine representative sequences for each clusters in a CDHIT.clstr file')
parser.add_argument('infile', action='store', help='Path to the input CD-HIT file for all reads')
parser.add_argument('indexfile', action='store', help='Path to the index file for all reads')
parser.add_argument('--version', action='version', version='%(prog)s 1.0')
args = parser.parse_args()
# Run through all Clusters
cluster_gen = parse(args.infile, idx_file_path=args.indexfile, edit_dist=False, similarity_count=True)
# Slow, Only do this once.
seqrec_lookup = SeqIO.index_db(args.indexfile)
rep_seq_correction_count = 0
clusters_count = 0
with open('clusters.temp', 'wb') as clust_file:
for cluster in cluster_gen:
# if fraction of reads 100% similar to representative seq is in majority,
# skip and write to new file
seq_similarity_ranking = cluster.editdist_counter.most_common()
print seq_similarity_ranking[0][0]
# Setup parser
parser = argparse.ArgumentParser(description='Procedure to write FastQ output file for clusters in a CDHIT.clstr file')
parser.add_argument('indexfile', action='store', help='Path to the index file for all reads')
parser.add_argument('ouputdir', action='store', help='Directory where outputs are stored to the index file for all reads')
parser.add_argument('--version', action='version', version='%(prog)s 1.0')
args = parser.parse_args()
if not os.path.exists(args.ouputdir):
os.makedirs(args.ouputdir)
db = SeqIO.index_db(args.indexfilepath)
idx_file_dir = os.path.split(args.indexfilepath)[0]
cluster_gen = parse(sys.stdin, idx_file_path=args.indexfile, edit_dist=False)
size_counter = Counter()
count = 0
for cluster in cluster_gen:
count += 1
# Write to output file
size_counter[str(cluster.size)] += 1
# Get the sequence records for the cluster
seqs = []
if os.getcwd() != idx_file_dir: