def parse_alignments(contigs_fpaths, contig_report_fpath_pattern):
lists_of_aligned_blocks = []
for contigs_fpath in contigs_fpaths:
if contig_report_fpath_pattern:
report_fpath = contig_report_fpath_pattern % qutils.label_from_fpath_for_fname(
contigs_fpath)
aligned_blocks, misassembled_id_to_structure = parse_aligner_contig_report(
report_fpath)
if aligned_blocks is None:
continue
aligned_blocks = check_misassembled_blocks(
aligned_blocks,
misassembled_id_to_structure,
filter_local=True)
lists_of_aligned_blocks.append(aligned_blocks)
if lists_of_aligned_blocks:
max_contigs = max([
len(aligned_blocks) for aligned_blocks in lists_of_aligned_blocks
])
return get_assemblies(contigs_fpaths,
lists_of_aligned_blocks).assemblies, max_contigs
else:
return None, None
def parse_alignments(contigs_fpaths, contig_report_fpath_pattern):
lists_of_aligned_blocks = []
for contigs_fpath in contigs_fpaths:
if contig_report_fpath_pattern:
report_fpath = contig_report_fpath_pattern % qutils.label_from_fpath_for_fname(contigs_fpath)
aligned_blocks, misassembled_id_to_structure = parse_nucmer_contig_report(report_fpath)
if aligned_blocks is None:
continue
aligned_blocks = check_misassembled_blocks(aligned_blocks, misassembled_id_to_structure, filter_local=True)
lists_of_aligned_blocks.append(aligned_blocks)
if lists_of_aligned_blocks:
max_contigs = max([len(aligned_blocks) for aligned_blocks in lists_of_aligned_blocks])
return get_assemblies(contigs_fpaths, lists_of_aligned_blocks).assemblies, max_contigs
else:
return None, None
def do(contigs_fpaths,
contig_report_fpath_pattern,
output_dirpath,
ref_fpath,
cov_fpath=None,
physical_cov_fpath=None,
gc_fpath=None,
stdout_pattern=None,
find_similar=True,
features=None,
json_output_dir=None,
genes_by_labels=None):
make_output_dir(output_dirpath)
lists_of_aligned_blocks = []
contigs_by_assemblies = OrderedDict()
structures_by_labels = {}
ambiguity_alignments_by_labels = {}
total_genome_size = 0
reference_chromosomes = OrderedDict()
contig_names_by_refs = None
assemblies = None
chr_names = []
features_data = None
plot_fpath = None
if ref_fpath:
for name, seq in fastaparser.read_fasta(ref_fpath):
chr_name = name.split()[0]
chr_names.append(chr_name)
chr_len = len(seq)
total_genome_size += chr_len
reference_chromosomes[chr_name] = chr_len
virtual_genome_shift = 100
sorted_ref_names = sorted(reference_chromosomes,
key=reference_chromosomes.get,
reverse=True)
sorted_ref_lengths = sorted(reference_chromosomes.values(),
reverse=True)
cumulative_ref_lengths = [0]
if ref_labels_by_chromosomes:
contig_names_by_refs = ref_labels_by_chromosomes
elif sum(reference_chromosomes.values()
) > qconfig.MAX_SIZE_FOR_COMB_PLOT:
contig_names_by_refs = dict()
if len(chr_names) > qconfig.ICARUS_MAX_CHROMOSOMES:
summary_len = 0
num_parts = 1
html_name = qconfig.alignment_viewer_part_name + str(num_parts)
for chr_name, chr_len in reference_chromosomes.items():
summary_len += chr_len
contig_names_by_refs[chr_name] = html_name
if summary_len >= qconfig.MAX_SIZE_FOR_COMB_PLOT:
summary_len = 0
num_parts += 1
html_name = qconfig.alignment_viewer_part_name + str(
num_parts)
else:
for chr_name in chr_names:
contig_names_by_refs[chr_name] = chr_name
for i, chr in enumerate(chr_names):
chr_length = reference_chromosomes[chr]
len_to_append = cumulative_ref_lengths[-1] + chr_length
if contig_names_by_refs:
if i < len(chr_names) - 1 and contig_names_by_refs[
chr] != contig_names_by_refs[chr_names[i + 1]]:
len_to_append = 0
cumulative_ref_lengths.append(len_to_append)
virtual_genome_size = sum(reference_chromosomes.values(
)) + virtual_genome_shift * (len(reference_chromosomes.values()) - 1)
for contigs_fpath in contigs_fpaths:
label = qconfig.assembly_labels_by_fpath[contigs_fpath]
if not contig_report_fpath_pattern:
contigs = parse_contigs_fpath(contigs_fpath)
else:
report_fpath = contig_report_fpath_pattern % qutils.label_from_fpath_for_fname(
contigs_fpath)
aligned_blocks, misassembled_id_to_structure, contigs, ambiguity_alignments = parse_aligner_contig_report(
report_fpath, list(reference_chromosomes.keys()),
cumulative_ref_lengths)
if not contigs:
contigs = parse_contigs_fpath(contigs_fpath)
if aligned_blocks is None:
return None
for block in aligned_blocks:
block.label = label
aligned_blocks = check_misassembled_blocks(
aligned_blocks, misassembled_id_to_structure)
lists_of_aligned_blocks.append(aligned_blocks)
structures_by_labels[label] = misassembled_id_to_structure
if qconfig.ambiguity_usage == 'all':
ambiguity_alignments_by_labels[label] = ambiguity_alignments
contigs_by_assemblies[label] = contigs
if ref_fpath:
features_data = parse_features_data(features, cumulative_ref_lengths,
chr_names)
if contigs_fpaths and qconfig.gene_finding:
parse_genes_data(contigs_by_assemblies, genes_by_labels)
if reference_chromosomes and lists_of_aligned_blocks:
assemblies = get_assemblies(contigs_fpaths, lists_of_aligned_blocks,
virtual_genome_size, find_similar)
if qconfig.draw_svg:
plot_fpath = draw_alignment_plot(assemblies, virtual_genome_size,
output_dirpath, sorted_ref_names,
sorted_ref_lengths,
virtual_genome_shift)
if (assemblies or contigs_by_assemblies) and qconfig.create_icarus_html:
icarus_html_fpath = js_data_gen(
assemblies,
contigs_fpaths,
reference_chromosomes,
output_dirpath,
structures_by_labels,
contig_names_by_refs=contig_names_by_refs,
ref_fpath=ref_fpath,
stdout_pattern=stdout_pattern,
ambiguity_alignments_by_labels=ambiguity_alignments_by_labels,
contigs_by_assemblies=contigs_by_assemblies,
features_data=features_data,
gc_fpath=gc_fpath,
cov_fpath=cov_fpath,
physical_cov_fpath=physical_cov_fpath,
json_output_dir=json_output_dir)
else:
icarus_html_fpath = None
return icarus_html_fpath, plot_fpath
def do(contigs_fpaths, contig_report_fpath_pattern, output_dirpath, ref_fpath, cov_fpath=None, physical_cov_fpath=None,
stdout_pattern=None, find_similar=True, features=None, json_output_dir=None, genes_by_labels=None):
make_output_dir(output_dirpath)
lists_of_aligned_blocks = []
contigs_by_assemblies = OrderedDict()
structures_by_labels = {}
ambiguity_alignments_by_labels = {}
total_genome_size = 0
reference_chromosomes = OrderedDict()
contig_names_by_refs = None
assemblies = None
chr_names = []
features_data = None
plot_fpath = None
max_small_chromosomes = 10
if ref_fpath:
for name, seq in fastaparser.read_fasta(ref_fpath):
chr_name = name.split()[0]
chr_names.append(chr_name)
chr_len = len(seq)
total_genome_size += chr_len
reference_chromosomes[chr_name] = chr_len
virtual_genome_shift = 100
sorted_ref_names = sorted(reference_chromosomes, key=reference_chromosomes.get, reverse=True)
sorted_ref_lengths = sorted(reference_chromosomes.values(), reverse=True)
cumulative_ref_lengths = [0]
if ref_labels_by_chromosomes:
contig_names_by_refs = ref_labels_by_chromosomes
elif sum(reference_chromosomes.values()) > qconfig.MAX_SIZE_FOR_COMB_PLOT:
contig_names_by_refs = dict()
if len(chr_names) > max_small_chromosomes:
summary_len = 0
num_parts = 1
html_name = qconfig.alignment_viewer_part_name + str(num_parts)
for chr_name, chr_len in reference_chromosomes.items():
summary_len += chr_len
contig_names_by_refs[chr_name] = html_name
if summary_len >= qconfig.MAX_SIZE_FOR_COMB_PLOT:
summary_len = 0
num_parts += 1
html_name = qconfig.alignment_viewer_part_name + str(num_parts)
else:
for chr_name in chr_names:
contig_names_by_refs[chr_name] = chr_name
for i, chr in enumerate(chr_names):
chr_length = reference_chromosomes[chr]
len_to_append = cumulative_ref_lengths[-1] + chr_length
if contig_names_by_refs:
if i < len(chr_names) - 1 and contig_names_by_refs[chr] != contig_names_by_refs[chr_names[i + 1]]:
len_to_append = 0
cumulative_ref_lengths.append(len_to_append)
virtual_genome_size = sum(reference_chromosomes.values()) + virtual_genome_shift * (len(reference_chromosomes.values()) - 1)
for contigs_fpath in contigs_fpaths:
label = qconfig.assembly_labels_by_fpath[contigs_fpath]
if not contig_report_fpath_pattern:
contigs = parse_contigs_fpath(contigs_fpath)
else:
report_fpath = contig_report_fpath_pattern % qutils.label_from_fpath_for_fname(contigs_fpath)
aligned_blocks, misassembled_id_to_structure, contigs, ambiguity_alignments = parse_nucmer_contig_report(report_fpath,
list(reference_chromosomes.keys()), cumulative_ref_lengths)
if not contigs:
contigs = parse_contigs_fpath(contigs_fpath)
if aligned_blocks is None:
return None
for block in aligned_blocks:
block.label = label
aligned_blocks = check_misassembled_blocks(aligned_blocks, misassembled_id_to_structure)
lists_of_aligned_blocks.append(aligned_blocks)
structures_by_labels[label] = misassembled_id_to_structure
if qconfig.ambiguity_usage == 'all':
ambiguity_alignments_by_labels[label] = ambiguity_alignments
contigs_by_assemblies[label] = contigs
if contigs_fpaths and ref_fpath and features:
features_data = parse_features_data(features, cumulative_ref_lengths, chr_names)
if contigs_fpaths and qconfig.gene_finding:
parse_genes_data(contigs_by_assemblies, genes_by_labels)
if reference_chromosomes and lists_of_aligned_blocks:
assemblies = get_assemblies(contigs_fpaths, virtual_genome_size, lists_of_aligned_blocks, find_similar)
if qconfig.draw_svg:
plot_fpath = draw_alignment_plot(assemblies, virtual_genome_size, output_dirpath, sorted_ref_names, sorted_ref_lengths, virtual_genome_shift)
if (assemblies or contigs_by_assemblies) and qconfig.create_icarus_html:
icarus_html_fpath = js_data_gen(assemblies, contigs_fpaths, reference_chromosomes,
output_dirpath, structures_by_labels, contig_names_by_refs=contig_names_by_refs, ref_fpath=ref_fpath, stdout_pattern=stdout_pattern,
ambiguity_alignments_by_labels=ambiguity_alignments_by_labels, contigs_by_assemblies=contigs_by_assemblies,
features_data=features_data, cov_fpath=cov_fpath, physical_cov_fpath=physical_cov_fpath, json_output_dir=json_output_dir)
else:
icarus_html_fpath = None
return icarus_html_fpath, plot_fpath