def loadFromFile(self, fname, num_names=True):
# type: (str, bool) -> ContigCollection
for rec in SeqIO.parse_by_name(fname):
if num_names:
self.add(
Contig(rec.seq,
str(basic.parseNegativeNumberAndMod(rec.id))))
else:
self.add(Contig(rec.seq, rec.id))
return self
def main(args):
flye_dir = sys.argv[1]
repeats, starts, ends = parse(sys.argv[2])
graph_file, unique_file, disjointigs_file, rep_dir, tmp, their_file = cl_params.parseFlyeDir(
flye_dir)
dump = os.path.join(rep_dir, "read_alignment_dump")
reads_file = sys.argv[3]
dir = sys.argv[4]
CreateLog(dir)
print " ".join(args)
print "Printing contigs"
edges_file = os.path.join(rep_dir, "graph_before_rr.fasta")
edges = ContigStorage().loadFromFasta(open(edges_file, "r"))
unique = open(os.path.join(dir, "contigs"), "w")
for l in starts:
seq = "".join(map(lambda eid: edges[eid].seq, l))
if len(seq) > 15000:
seq = seq[-15000:]
SeqIO.write(NamedSequence(seq, "(" + "_".join(l) + ")"), unique,
"fasta")
for l in ends:
seq = "".join(map(lambda eid: edges[eid].seq, l))
if len(seq) > 15000:
seq = seq[:15000]
SeqIO.write(NamedSequence(basic.RC(seq), "(" + "_".join(l) + ")"),
unique, "fasta")
unique.close()
print "Selecting reads"
reads = set()
cur_read = None
als = []
for s in open(dump).readlines():
if s.startswith("Chain"):
if cur_read == "06dfd536-e258-446f-a019-8d8340ef831e":
print als
for al in als:
if al in repeats:
if cur_read == "06dfd536-e258-446f-a019-8d8340ef831e":
print "oppa"
reads.add(cur_read)
break
als = []
else:
s = s.split()
cur_read = s[2][1:]
eid = s[6].split("_")[1]
if s[6][0] == "-":
eid = "-" + eid
if cur_read == "06dfd536-e258-446f-a019-8d8340ef831e":
print eid
als.append(eid)
print "Selected", len(reads), "reads"
print "\n".join(reads)
print "Reading and printing reads"
freads = open(os.path.join(dir, "reads.fasta"), "w")
cnt = 0
for read in SeqIO.parse_by_name(reads_file):
cnt += 1
if cnt % 10000 == 0:
print cnt
if read.id in reads:
SeqIO.write(read, freads, "fasta")
freads.close()
def main(flye_dir, rf, dir, edge_id, to_resolve, min_contig_length):
params.technology = "nano"
basic.ensure_dir_existance(dir)
basic.CreateLog(dir)
dd = DirDistributor(os.path.join(dir, "alignments"))
aligner = Aligner(dd)
print " ".join(sys.argv)
print "Reading graph"
graph = SimpleGraph().ReadDot(
os.path.join(flye_dir, "20-repeat", "graph_before_rr.gv"))
graph.FillSeq(os.path.join(flye_dir, "20-repeat", "graph_before_rr.fasta"),
True)
print "Extracting relevant graph component"
edge_ids = edge_id.split(",")
to_resolve = to_resolve.split(",")
to_resolve = [(a, int(b))
for a, b in zip(to_resolve[0::2], to_resolve[1::2])]
unique = uniqueNeighbours(edge_ids, graph, min_contig_length)
if rf == "none":
return
print "Finding reads that align to", edge_ids
reads_to_resolve = dict() # type: Dict[str, List[str]]
for eid, mult in to_resolve:
reads_to_resolve[eid] = []
for unique_edge, initial in unique:
reads_to_resolve[initial] = []
relevant_read_ids = set()
for rid, eid in parseReadDump(
os.path.join(flye_dir, "20-repeat", "read_alignment_dump")):
if eid in edge_ids:
relevant_read_ids.add(rid)
print rid, eid
for rid, eid in parseReadDump(
os.path.join(flye_dir, "20-repeat", "read_alignment_dump")):
if rid in relevant_read_ids and eid in reads_to_resolve:
reads_to_resolve[eid].append(rid)
for eid in reads_to_resolve:
reads_to_resolve[eid] = list(set(reads_to_resolve[eid]))
print "Reading reads"
res_reads = ContigStorage()
res = open(os.path.join(dir, "reads.fasta"), "w")
for read in SeqIO.parse_by_name(rf):
if read.id in relevant_read_ids:
res_reads.add(Contig(read.seq, read.id))
SeqIO.write(read, res, "fasta")
res.close()
random_down = open(os.path.join(dir, "random_down.fasta"), "w")
cnt = 0
for read in res_reads:
if cnt % 5 == 0:
SeqIO.write(read, random_down, "fasta")
cnt += 1
random_down.close()
res = open(os.path.join(dir, "contigs.fasta"), "w")
lcf = open(os.path.join(dir, "contigs.lc"), "w")
for eid, mult in to_resolve:
repeat_reads = [res_reads[rid] for rid in reads_to_resolve[eid]]
print reads_to_resolve[eid]
print map(str, repeat_reads)
split_contigs = splitRepeat(aligner, graph.e[eid].seq, mult,
repeat_reads, min_contig_length)
if split_contigs is None:
print "Failed to resove edge", eid, "Aborting"
print "Edge", eid, "was split into", mult, "copies"
for contig, contig_reads in split_contigs:
print contig.id
SeqIO.write(contig, res, "fasta")
lcf.write(contig.id + "\n")
lcf.write(" ".join([r.id for r in contig_reads]) + "\n")
res = open(os.path.join(dir, "contigs.fasta"), "w")
for unique_edge, initial in unique:
print unique_edge.id
SeqIO.write(unique_edge, res, "fasta")
lcf.write(unique_edge.id + "\n")
lcf.write(" ".join(reads_to_resolve[initial]) + "\n")
res.close()