def _run_polisher_only(args):
"""
Runs standalone polisher
"""
logger.info("Running Flye polisher")
logger.debug("Cmd: %s", " ".join(sys.argv))
bam_input = False
for read_file in args.reads:
if not os.path.exists(read_file):
raise ResumeException("Can't open " + read_file)
without_gz = read_file.rstrip(".gz")
if not any([
without_gz.endswith(x)
for x in ["fasta", "fa", "fastq", "fq", "bam"]
]):
raise ResumeException(
"Unsupported input. Supported types: fasta/fastq/bam")
if without_gz.endswith("bam"):
bam_input = True
if bam_input and len(args.reads) > 1:
raise ResumeException("Only single bam input supported")
pol.polish(args.polish_target,
args.reads,
args.out_dir,
args.num_iters,
args.threads,
args.platform,
args.read_type,
output_progress=True)
logger.info("Done!")
def _run_polisher_only(args):
"""
Runs standalone polisher
"""
logger.info("Running Flye polisher")
logger.debug("Cmd: {0}".format(" ".join(sys.argv)))
pol.polish(args.polish_target, args.reads, args.out_dir,
args.num_iters, args.threads, args.platform,
output_progress=True)
def run(self):
super(JobPolishing, self).run()
if not os.path.isdir(self.polishing_dir):
os.mkdir(self.polishing_dir)
pol.polish(self.in_contigs, self.args.reads, self.polishing_dir,
self.args.num_iters, self.args.threads, self.args.platform,
output_progress=True)
polished_file = os.path.join(self.polishing_dir, "polished_{0}.fasta"
.format(self.args.num_iters))
pol.generate_polished_edges(self.in_graph_edges, self.in_graph_gfa,
polished_file,
self.polishing_dir, self.args.platform,
self.args.threads)
def run(self):
super(JobPolishing, self).run()
if not os.path.isdir(self.polishing_dir):
os.mkdir(self.polishing_dir)
contigs, stats = \
pol.polish(self.in_contigs, self.args.reads, self.polishing_dir,
self.args.num_iters, self.args.threads, self.args.platform,
output_progress=True)
#contigs = os.path.join(self.polishing_dir, "polished_1.fasta")
#stats = os.path.join(self.polishing_dir, "contigs_stats.txt")
pol.filter_by_coverage(self.args, stats, contigs,
self.out_files["stats"], self.out_files["contigs"])
pol.generate_polished_edges(self.in_graph_edges, self.in_graph_gfa,
self.out_files["contigs"],
self.polishing_dir, self.args.platform,
self.args.threads)
os.remove(contigs)
def run(self):
super(JobPolishing, self).run()
if not os.path.isdir(self.polishing_dir):
os.mkdir(self.polishing_dir)
contigs, stats = \
pol.polish(self.in_contigs, self.args.reads, self.polishing_dir,
self.args.num_iters, self.args.threads, self.args.platform,
self.args.read_type, output_progress=True)
#contigs = os.path.join(self.polishing_dir, "polished_1.fasta")
#stats = os.path.join(self.polishing_dir, "contigs_stats.txt")
pol.filter_by_coverage(self.args, stats, contigs,
self.out_files["stats"],
self.out_files["contigs"])
pol.generate_polished_edges(self.in_graph_edges, self.in_graph_gfa,
self.out_files["contigs"],
self.polishing_dir, self.args.platform,
stats, self.args.threads)
os.remove(contigs)
if os.path.getsize(self.out_files["contigs"]) == 0:
raise asm.AssembleException("No contigs were assembled - "
"pipeline stopped")
def assemble_short_plasmids(args, work_dir, contigs_path):
logger.debug("Extracting unmapped reads")
reads2contigs_mapping = os.path.join(work_dir, "reads2contigs.paf")
make_alignment(contigs_path, args.reads, args.threads,
work_dir, args.platform, reads2contigs_mapping,
reference_mode=True, sam_output=False)
unmapped_reads_path = os.path.join(work_dir, "unmapped_reads.fasta")
unmapped.extract_unmapped_reads(args, reads2contigs_mapping,
unmapped_reads_path,
mapping_rate_threshold=0.5)
logger.debug("Finding self-mappings for unmapped reads")
unmapped_reads_mapping = os.path.join(work_dir, "unmapped_ava.paf")
make_alignment(unmapped_reads_path, [unmapped_reads_path], args.threads,
work_dir, args.platform, unmapped_reads_mapping,
reference_mode=False, sam_output=False)
logger.debug("Extracting circular reads")
circular_reads = circular.extract_circular_reads(unmapped_reads_mapping)
logger.debug("Extracted %d circular reads", len(circular_reads))
logger.debug("Extracing circular pairs")
circular_pairs = circular.extract_circular_pairs(unmapped_reads_mapping)
logger.debug("Extracted %d circular pairs", len(circular_pairs))
#extracting only the necesssary subset of reads (the entire file could be pretty big)
interesting_reads = {}
for read in circular_reads:
interesting_reads[read] = None
for pair in circular_pairs:
interesting_reads[pair[0].query] = None
interesting_reads[pair[0].target] = None
for hdr, seq in fp.stream_sequence(unmapped_reads_path):
if hdr in interesting_reads:
interesting_reads[hdr] = seq
trimmed_circular_reads = \
circular.trim_circular_reads(circular_reads, interesting_reads)
trimmed_circular_pairs = \
circular.trim_circular_pairs(circular_pairs, interesting_reads)
trimmed_sequences_path = os.path.join(work_dir, "trimmed_sequences.fasta")
fp.write_fasta_dict(dict(list(trimmed_circular_reads.items()) +
list(trimmed_circular_pairs.items())),
trimmed_sequences_path)
logger.debug("Clustering circular sequences")
trimmed_sequences_mapping = os.path.join(work_dir, "trimmed.paf")
make_alignment(trimmed_sequences_path, [trimmed_sequences_path], args.threads,
work_dir, args.platform, trimmed_sequences_mapping,
reference_mode=False, sam_output=False)
plasmids = \
circular.extract_unique_plasmids(trimmed_sequences_mapping,
trimmed_sequences_path)
plasmids_raw = os.path.join(work_dir, "plasmids_raw.fasta")
fp.write_fasta_dict(plasmids, plasmids_raw)
_, polished_stats = \
pol.polish(plasmids_raw, [unmapped_reads_path], work_dir, 1,
args.threads, args.platform, output_progress=False)
#extract coverage
plasmids_with_coverage = {}
if os.path.isfile(polished_stats):
with open(polished_stats, "r") as f:
for line in f:
if line.startswith("#"): continue
tokens = line.strip().split()
seq_id, coverage = tokens[0], int(tokens[2])
if coverage > 0:
plasmids_with_coverage[seq_id] = plasmids[seq_id], coverage
logger.info("Added %d extra contigs", len(plasmids_with_coverage))
# remove all unnecesarry files
os.remove(reads2contigs_mapping)
os.remove(unmapped_reads_path)
os.remove(unmapped_reads_mapping)
os.remove(trimmed_sequences_path)
os.remove(trimmed_sequences_mapping)
return plasmids_with_coverage
def assemble_short_plasmids(args, work_dir, contigs_path):
logger.debug("Assembling short plasmids")
reads2contigs_mapping = os.path.join(work_dir, "reads2contigs.paf")
make_alignment(contigs_path,
args.reads,
args.threads,
work_dir,
args.platform,
reads2contigs_mapping,
reference_mode=True,
sam_output=False)
logger.debug("Extracting unmapped reads")
unmapped_reads, n_processed_reads = \
unmapped.extract_unmapped_reads(args, reads2contigs_mapping,
mapping_rate_threshold=0.5)
n_unmapped_reads = len(unmapped_reads)
unmapped_reads_ratio = 100 * float(len(unmapped_reads)) / n_processed_reads
unmapped_reads_ratio = round(unmapped_reads_ratio, 1)
logger.debug("Extracted {} unmapped reads ({} %)".format(
n_unmapped_reads, unmapped_reads_ratio))
unmapped_reads_path = os.path.join(work_dir, "unmapped_reads.fasta")
fp.write_fasta_dict(unmapped_reads, unmapped_reads_path)
unmapped_reads_mapping = os.path.join(work_dir, "unmapped_ava.paf")
logger.debug("Finding self-mappings for unmapped reads")
make_alignment(unmapped_reads_path, [unmapped_reads_path],
args.threads,
work_dir,
args.platform,
unmapped_reads_mapping,
reference_mode=False,
sam_output=False)
logger.debug("Extracting circular reads")
circular_reads = circular.extract_circular_reads(unmapped_reads_mapping)
logger.debug("Extracted {} circular reads".format(len(circular_reads)))
logger.debug("Extracing circular pairs")
circular_pairs = circular.extract_circular_pairs(unmapped_reads_mapping)
logger.debug("Extracted {} circular pairs".format(len(circular_pairs)))
logger.debug("Extracting unique plasmids from circular sequences")
trimmed_circular_reads = \
circular.trim_circular_reads(circular_reads, unmapped_reads)
trimmed_circular_pairs = \
circular.trim_circular_pairs(circular_pairs, unmapped_reads)
trimmed_sequences_path = os.path.join(work_dir, "trimmed_sequences.fasta")
fp.write_fasta_dict(
dict(trimmed_circular_reads.items() + trimmed_circular_pairs.items()),
trimmed_sequences_path)
trimmed_sequences_mapping = os.path.join(work_dir, "trimmed.paf")
make_alignment(trimmed_sequences_path, [trimmed_sequences_path],
args.threads,
work_dir,
args.platform,
trimmed_sequences_mapping,
reference_mode=False,
sam_output=False)
plasmids = \
circular.extract_unique_plasmids(trimmed_sequences_mapping,
trimmed_sequences_path)
plasmids_raw = os.path.join(work_dir, "plasmids_raw.fasta")
fp.write_fasta_dict(plasmids, plasmids_raw)
pol.polish(plasmids_raw, [unmapped_reads_path],
work_dir,
1,
args.threads,
args.platform,
output_progress=False)
#extract coverage
plasmids_with_coverage = {}
if os.path.isfile(os.path.join(work_dir, "contigs_stats.txt")):
with open(os.path.join(work_dir, "contigs_stats.txt"), "r") as f:
for line in f:
if line.startswith("seq"): continue
tokens = line.strip().split()
seq_id, coverage = tokens[0], int(tokens[2])
if coverage > 0:
plasmids_with_coverage[seq_id] = plasmids[seq_id], coverage
logger.info("Added {} extra contigs".format(len(plasmids_with_coverage)))
# remove all unnecesarry files
os.remove(reads2contigs_mapping)
os.remove(unmapped_reads_path)
os.remove(unmapped_reads_mapping)
os.remove(trimmed_sequences_path)
os.remove(trimmed_sequences_mapping)
return plasmids_with_coverage
