def do_job(nucmer_coords, debug_dir, circular, only_predicted):
used_contigs = os.path.join(debug_dir, "filtered_contigs.txt")
true_adj_out = os.path.join(debug_dir, "true_edges.dot")
base_dot = os.path.join(debug_dir, "breakpoint_graph.dot")
overlap_dot = os.path.join(debug_dir, "../contigs_overlap.dot")
predicted_dot = os.path.join(debug_dir, "predicted_edges.dot")
phylogeny_in = os.path.join(debug_dir, "phylogeny.txt")
phylogeny_out = os.path.join(debug_dir, "phylogeny.png")
draw_phylogeny(phylogeny_in, phylogeny_out)
contigs = get_contig_permutations(used_contigs)
if nucmer_coords != "-":
alignment = parse_nucmer_coords(nucmer_coords)
alignment = list(filter(lambda e: e.qry.seq_id in contigs, alignment))
#alignment = join_collinear(alignment)
alignment = filter_by_coverage(alignment, 0.7)
alignment = join_collinear(alignment)
break_contigs = verify_alignment(alignment, contigs)
true_adj = get_true_adjacencies(alignment, contigs,
break_contigs, circular)
else:
true_adj = []
output_edges(true_adj, true_adj_out)
g = compose_breakpoint_graph(base_dot, predicted_dot, true_adj)
if os.path.exists(overlap_dot):
add_overlap_edges(g, overlap_dot, used_contigs)
output_graph(g, debug_dir, only_predicted)
def do_job(nucmer_coords, debug_dir, circular, only_predicted):
used_contigs = os.path.join(debug_dir, "filtered_contigs.txt")
true_adj_out = os.path.join(debug_dir, "true_edges.dot")
base_dot = os.path.join(debug_dir, "breakpoint_graph.dot")
overlap_dot = os.path.join(debug_dir, "../contigs_overlap.dot")
predicted_dot = os.path.join(debug_dir, "predicted_edges.dot")
phylogeny_in = os.path.join(debug_dir, "phylogeny.txt")
phylogeny_out = os.path.join(debug_dir, "phylogeny.png")
draw_phylogeny(phylogeny_in, phylogeny_out)
contigs = get_contig_permutations(used_contigs)
if nucmer_coords != "-":
alignment = parse_nucmer_coords(nucmer_coords)
alignment = list([e for e in alignment if e.qry.seq_id in contigs])
#alignment = join_collinear(alignment)
alignment = filter_by_coverage(alignment, 0.7)
alignment = join_collinear(alignment)
break_contigs = verify_alignment(alignment, contigs)
true_adj = get_true_adjacencies(alignment, contigs, break_contigs,
circular)
else:
true_adj = []
output_edges(true_adj, true_adj_out)
g = compose_breakpoint_graph(base_dot, predicted_dot, true_adj)
if os.path.exists(overlap_dot):
add_overlap_edges(g, overlap_dot, used_contigs)
output_graph(g, debug_dir, only_predicted)
def do_job(nucmer_coords, scaffolds_ord):
alignment = parse_nucmer_coords(nucmer_coords)
alignment = join_collinear(alignment)
alignment = filter_by_coverage(alignment, 0.45)
entry_ord, chr_len, contig_len = get_order(alignment)
scaffolds = parse_contigs_order(scaffolds_ord)
total_breaks = 0
total_gaps = 0
total_contigs = 0
for s in scaffolds:
print("\n>" + s.name)
prev_aln = []
prev_strand = None
increasing = None
breaks = []
for contig in s.contigs:
miss_ord = False
miss_strand = False
#checking order
if prev_aln:
if increasing is not None:
if not agreement_ord(increasing, prev_aln,
entry_ord[contig.name], chr_len):
increasing = None
breaks.append(contig.name)
total_breaks += 1
miss_ord = True
elif len(entry_ord[contig.name]) == 1 and len(prev_aln) == 1:
increasing = (entry_ord[contig.name][0].index >
prev_aln[0].index)
#checking strand
cur_strand = [h.sign * contig.sign for h in entry_ord[contig.name]]
if not miss_ord and prev_strand and cur_strand:
if not agreement_strands(prev_strand, cur_strand, increasing):
breaks.append(contig.name)
total_breaks += 1
miss_strand = True
increasing = None
if gap_count(prev_aln, entry_ord[contig.name]) > 0:
total_gaps += 1
#only if this contig has alignments
if entry_ord[contig.name]:
prev_aln = entry_ord[contig.name]
prev_strand = cur_strand
#output
sign = "+" if contig.sign > 0 else "-"
pos_list = list(map(str, entry_ord[contig.name]))
pos_list_str = (str(pos_list) if len(pos_list) < 5 else
str(pos_list[:5]) + "...")
print("{0}{1}\t{2}\t{3}".format(sign, contig.name,
contig_len[contig.name],
pos_list_str),
end="")
print("\t<<<order" if miss_ord else "", end="")
print("\t<<<strand" if miss_strand else "", end="")
print("")
total_contigs += 1
###
print("\tmiss-ordered: ", len(breaks))
print("\nTotal miss-ordered:", total_breaks)
print("Total gaps:", total_gaps)
print("Total contigs:", total_contigs)
print("Total scaffolds:", len(scaffolds))
def do_job(nucmer_coords, scaffolds_ord):
alignment = parse_nucmer_coords(nucmer_coords)
alignment = join_collinear(alignment)
alignment = filter_by_coverage(alignment, 0.45)
entry_ord, chr_len, contig_len = get_order(alignment)
scaffolds = parse_contigs_order(scaffolds_ord)
total_breaks = 0
total_gaps = 0
total_contigs = 0
for s in scaffolds:
print("\n>" + s.name)
prev_aln = None
prev_strand = None
increasing = None
breaks = []
for contig in s.contigs:
miss_ord = False
miss_strand = False
#checking order
if prev_aln:
if increasing is not None:
if not agreement_ord(increasing, prev_aln,
entry_ord[contig.name], chr_len):
increasing = None
breaks.append(contig.name)
total_breaks += 1
miss_ord = True
elif len(entry_ord[contig.name]) == 1 and len(prev_aln) == 1:
increasing = (entry_ord[contig.name][0].index >
prev_aln[0].index)
#checking strand
cur_strand = list(map(lambda h: h.sign * contig.sign,
entry_ord[contig.name]))
if not miss_ord and prev_strand and cur_strand:
if not agreement_strands(prev_strand, cur_strand, increasing):
breaks.append(contig.name)
total_breaks += 1
miss_strand = True
increasing = None
if gap_count(prev_aln, entry_ord[contig.name]) > 0:
total_gaps += 1
#only if this contig has alignments
if entry_ord[contig.name]:
prev_aln = entry_ord[contig.name]
prev_strand = cur_strand
#output
sign = "+" if contig.sign > 0 else "-"
pos_list = list(map(str, entry_ord[contig.name]))
pos_list_str = (str(pos_list) if len(pos_list) < 5 else
str(pos_list[:5]) + "...")
print("{0}{1}\t{2}\t{3}".format(sign, contig.name,
contig_len[contig.name], pos_list_str),
end="")
print("\t<<<order" if miss_ord else "", end="")
print("\t<<<strand" if miss_strand else "", end="")
print("")
total_contigs += 1
###
print("\tmiss-ordered: ", len(breaks))
print("\nTotal miss-ordered:", total_breaks)
print("Total gaps:", total_gaps)
print("Total contigs:", total_contigs)
print("Total scaffolds:", len(scaffolds))