def align_misasm_broken(out_prefix):
current_path = os.getcwd()
os.chdir('ctg_alignments')
ctgs_file = out_prefix + ".misasm.break.fa"
cmd = '{} -k19 -w19 -t{} ../../{} {} ' \
'> contigs_brk_against_ref.paf 2> contigs_brk_against_ref.paf.log'.format(minimap_path, t, reference_file,
ctgs_file)
if not os.path.isfile('contigs_brk_against_ref.paf'):
run(cmd)
os.chdir(current_path)
def align_pms(m_path, num_threads, in_reference_file):
current_path = os.getcwd()
output_path = current_path + '/pm_alignments'
if not os.path.exists(output_path):
os.makedirs(output_path)
os.chdir('pm_alignments')
cmd = '{} -ax asm5 --cs -t{} ../../{} {} ' \
'> pm_against_ref.sam 2> pm_contigs_against_ref.sam.log'.format(m_path, num_threads,
in_reference_file, '../ragoo.fasta')
if not os.path.isfile('pm_against_ref.sam'):
run(cmd)
os.chdir(current_path)
def align_breaks(break_type, m_path, in_reference_file, in_contigs_file, in_num_threads):
current_path = os.getcwd()
os.chdir('chimera_break')
if break_type == 'inter':
cmd = '{} -k19 -w19 -t{} ../../{} {} ' \
'> inter_contigs_against_ref.paf 2> inter_contigs_against_ref.paf.log'.format(m_path, in_num_threads, in_reference_file, in_contigs_file)
if not os.path.isfile('inter_contigs_against_ref.paf'):
run(cmd)
else:
cmd = '{} -k19 -w19 -t{} ../../{} {} ' \
'> intra_contigs_against_ref.paf 2> intra_contigs_against_ref.paf.log'.format(m_path, in_num_threads, in_reference_file, in_contigs_file)
if not os.path.isfile('intra_contigs_against_ref.paf'):
run(cmd)
os.chdir(current_path)
def align_pms(m_path, num_threads, in_reference_file, args):
current_path = os.getcwd()
output_path = os.path.join(current_path, 'pm_alignments')
if not os.path.exists(output_path):
os.makedirs(output_path)
assert os.path.exists("ragoo.fasta")
query = os.path.abspath("ragoo.fasta")
os.chdir('pm_alignments')
cmd = '{} -ax asm5 --cs -t{} -I {} {} {} ' \
'> pm_against_ref.sam 2> pm_contigs_against_ref.sam.log'.format(m_path, num_threads, args.I,
os.path.relpath(in_reference_file),
os.path.relpath(query))
if not os.path.isfile('pm_against_ref.sam'):
run(cmd)
os.chdir(current_path)
def get_SVs(sv_min, sv_max, in_ref_file):
current_path = os.getcwd()
os.chdir('pm_alignments')
# Change this when setup.py is ready. Just call script directly
cmd = 'sam2delta.py pm_against_ref.sam'
if not os.path.isfile('pm_against_ref.sam.delta'):
run(cmd)
cmd_2 = 'Assemblytics_uniq_anchor.py --delta pm_against_ref.sam.delta --unique-length 10000 --out assemblytics_out --keep-small-uniques'
if not os.path.isfile(
'assemblytics_out.Assemblytics.unique_length_filtered_l10000.delta'
):
run(cmd_2)
cmd_3 = 'Assemblytics_between_alignments.pl assemblytics_out.coords.tab %r %r all-chromosomes exclude-longrange bed > assemblytics_out.variants_between_alignments.bed' % (
sv_min, sv_max)
if not os.path.isfile('assemblytics_out.variants_between_alignments.bed'):
run(cmd_3)
cmd_4 = 'Assemblytics_within_alignment.py --delta assemblytics_out.Assemblytics.unique_length_filtered_l10000.delta --min %r > assemblytics_out.variants_within_alignments.bed' % (
sv_min)
if not os.path.isfile('assemblytics_out.variants_within_alignments.bed'):
run(cmd_4)
header = "reference\tref_start\tref_stop\tID\tsize\tstrand\ttype\tref_gap_size\tquery_gap_size\tquery_coordinates\tmethod\n"
with open('assemblytics_out.variants_between_alignments.bed', 'r') as f1:
b1 = f1.read()
with open('assemblytics_out.variants_within_alignments.bed', 'r') as f2:
b2 = f2.read()
with open('assemblytics_out.Assemblytics_structural_variants.bed',
'w') as f:
f.write(header)
# Might need to add newlines here
f.write(b1)
f.write(b2)
# Filter out SVs caused by gaps
cmd_5 = 'filter_gap_SVs.py %s' % (in_ref_file)
run(cmd_5)
os.chdir(current_path)
def align_reads(m_path, num_threads, in_ctg_file, reads, tech='ont'):
current_path = os.getcwd()
output_path = current_path + '/ctg_alignments'
if not os.path.exists(output_path):
os.makedirs(output_path)
os.chdir('ctg_alignments')
if tech == 'sr':
cmd = '{} -x sr -t{} ../../{} ../../{} ' \
'> reads_against_ctg.paf 2> reads_against_ctg.paf.log'.format(m_path, num_threads, in_ctg_file, reads)
elif tech == 'corr':
cmd = '{} -x asm10 -t{} ../../{} ../../{} ' \
'> reads_against_ctg.paf 2> reads_against_ctg.paf.log'.format(m_path, num_threads, in_ctg_file, reads)
else:
raise ValueError("Only 'sr' or 'corr' are accepted for read type.")
if not os.path.isfile('reads_against_ctg.paf'):
run(cmd)
os.chdir(current_path)
with open(skip_file) as f:
for line in f:
skip_ctg.append(line.rstrip())
current_path = os.getcwd()
output_path = current_path + '/ragoo_output'
if not os.path.exists(output_path):
os.makedirs(output_path)
os.chdir(output_path)
# Run minimap2
cmd = '{} -k19 -w19 -t{} ../{} ../{} ' \
'> contigs_against_ref.paf 2> contigs_against_ref.paf.log'.format(minimap_path, t, reference_file, contigs_file)
if not os.path.isfile('contigs_against_ref.paf'):
run(cmd)
# Read in the minimap2 alignments just generated
log('-- Reading alignments')
alns = read_paf_alignments('contigs_against_ref.paf')
alns = clean_alignments(alns, l=1000, in_exclude_file=exclude_file)
# Process the gff file
if gff_file:
log('-- Getting gff features')
features = defaultdict(list)
z = GFFReader('../' + gff_file)
for i in z.parse_gff():
features[i.seqname].append(i)
# Break chimeras if desired
