def correct_meta_references(ref_fpaths,
corrected_dirpath,
downloaded_refs=False):
corrected_ref_fpaths = []
combined_ref_fpath = os.path.join(corrected_dirpath,
qconfig.combined_ref_name)
chromosomes_by_refs = {}
def _proceed_seq(seq_name, seq, ref_name, ref_fasta_ext, total_references,
ref_fpath):
seq_fname = ref_name
seq_fname += ref_fasta_ext
if total_references > 1:
corr_seq_fpath = corrected_ref_fpaths[-1]
else:
corr_seq_fpath = qutils.unique_corrected_fpath(
os.path.join(corrected_dirpath, seq_fname))
corrected_ref_fpaths.append(corr_seq_fpath)
corr_seq_name = qutils.name_from_fpath(corr_seq_fpath) + '_' + seq_name
if not qconfig.no_check:
corr_seq = correct_seq(seq, ref_fpath)
if not corr_seq:
return None, None
fastaparser.write_fasta(corr_seq_fpath, [(corr_seq_name, seq)], 'a')
contigs_analyzer.ref_labels_by_chromosomes[
corr_seq_name] = qutils.name_from_fpath(corr_seq_fpath)
chromosomes_by_refs[ref_name].append((corr_seq_name, len(seq)))
return corr_seq_name, corr_seq_fpath
ref_fnames = [os.path.basename(ref_fpath) for ref_fpath in ref_fpaths]
ref_names = []
for ref_fname in ref_fnames:
ref_name, ref_fasta_ext = qutils.splitext_for_fasta_file(ref_fname)
ref_names.append(ref_name)
excluded_ref_fpaths = []
ref_names = qutils.process_labels(ref_fpaths)
for ref_fpath, ref_name in zip(ref_fpaths, ref_names):
total_references = 0
ref_fname = os.path.basename(ref_fpath)
_, ref_fasta_ext = qutils.splitext_for_fasta_file(ref_fname)
chromosomes_by_refs[ref_name] = []
used_seq_names = defaultdict(int)
corr_seq_fpath = None
for i, (seq_name, seq) in enumerate(fastaparser.read_fasta(ref_fpath)):
total_references += 1
seq_name = correct_name(seq_name,
qutils.MAX_CONTIG_NAME - len(ref_name) - 1)
uniq_seq_name = get_uniq_name(seq_name, used_seq_names)
used_seq_names[seq_name] += 1
corr_seq_name, corr_seq_fpath = _proceed_seq(
uniq_seq_name, seq, ref_name, ref_fasta_ext, total_references,
ref_fpath)
if not corr_seq_name:
break
if corr_seq_fpath:
logger.main_info(' ' + ref_fpath + ' ==> ' +
qutils.name_from_fpath(corr_seq_fpath) + '')
fastaparser.write_fasta(combined_ref_fpath,
fastaparser.read_fasta(corr_seq_fpath),
'a')
elif downloaded_refs:
logger.warning(
'Skipping ' + ref_fpath + ' because it'
' is empty or contains incorrect sequences (header-only or with non-ACGTN characters)!'
)
# cleaning
for corr_seq_name, _ in chromosomes_by_refs[ref_name]:
del contigs_analyzer.ref_labels_by_chromosomes[corr_seq_name]
del chromosomes_by_refs[ref_name]
corrected_ref_fpaths.pop()
excluded_ref_fpaths.append(ref_fpath)
else:
logger.error(
'Reference file ' + ref_fpath +
' is empty or contains incorrect sequences (header-only or with non-ACGTN characters)!',
exit_with_code=1)
for excluded in excluded_ref_fpaths:
ref_fpaths.remove(excluded)
if len(chromosomes_by_refs) > 0:
logger.main_info(' All references were combined in ' +
qconfig.combined_ref_name)
else:
logger.warning('All references were skipped!')
return corrected_ref_fpaths, combined_ref_fpath, chromosomes_by_refs, ref_fpaths
def correct_meta_references(ref_fpaths, corrected_dirpath, downloaded_refs=False):
corrected_ref_fpaths = []
combined_ref_fpath = os.path.join(corrected_dirpath, qconfig.combined_ref_name)
chromosomes_by_refs = {}
def _proceed_seq(seq_name, seq, ref_name, ref_fasta_ext, total_references, ref_fpath):
seq_fname = ref_name
seq_fname += ref_fasta_ext
if total_references > 1:
corr_seq_fpath = corrected_ref_fpaths[-1]
else:
corr_seq_fpath = qutils.unique_corrected_fpath(os.path.join(corrected_dirpath, seq_fname))
corrected_ref_fpaths.append(corr_seq_fpath)
corr_seq_name = qutils.name_from_fpath(corr_seq_fpath) + '_' + seq_name
if not qconfig.no_check:
corr_seq = correct_seq(seq, ref_fpath)
if not corr_seq:
return None, None
fastaparser.write_fasta(corr_seq_fpath, [(corr_seq_name, seq)], 'a')
contigs_analyzer.ref_labels_by_chromosomes[corr_seq_name] = qutils.name_from_fpath(corr_seq_fpath)
chromosomes_by_refs[ref_name].append((corr_seq_name, len(seq)))
return corr_seq_name, corr_seq_fpath
ref_fnames = [os.path.basename(ref_fpath) for ref_fpath in ref_fpaths]
ref_names = []
for ref_fname in ref_fnames:
ref_name, ref_fasta_ext = qutils.splitext_for_fasta_file(ref_fname)
ref_names.append(ref_name)
excluded_ref_fpaths = []
ref_names = qutils.process_labels(ref_fpaths)
for ref_fpath, ref_name in zip(ref_fpaths, ref_names):
total_references = 0
ref_fname = os.path.basename(ref_fpath)
_, ref_fasta_ext = qutils.splitext_for_fasta_file(ref_fname)
chromosomes_by_refs[ref_name] = []
used_seq_names = defaultdict(int)
corr_seq_fpath = None
for i, (seq_name, seq) in enumerate(fastaparser.read_fasta(ref_fpath)):
total_references += 1
seq_name = correct_name(seq_name, qutils.MAX_CONTIG_NAME - len(ref_name) - 1)
uniq_seq_name = get_uniq_name(seq_name, used_seq_names)
used_seq_names[seq_name] += 1
corr_seq_name, corr_seq_fpath = _proceed_seq(uniq_seq_name, seq, ref_name, ref_fasta_ext, total_references, ref_fpath)
if not corr_seq_name:
break
if corr_seq_fpath:
logger.main_info(' ' + ref_fpath + ' ==> ' + qutils.name_from_fpath(corr_seq_fpath) + '')
fastaparser.write_fasta(combined_ref_fpath, fastaparser.read_fasta(corr_seq_fpath), 'a')
elif downloaded_refs:
logger.warning('Skipping ' + ref_fpath + ' because it'
' is empty or contains incorrect sequences (header-only or with non-ACGTN characters)!')
# cleaning
for corr_seq_name, _ in chromosomes_by_refs[ref_name]:
del contigs_analyzer.ref_labels_by_chromosomes[corr_seq_name]
del chromosomes_by_refs[ref_name]
corrected_ref_fpaths.pop()
excluded_ref_fpaths.append(ref_fpath)
else:
logger.error('Reference file ' + ref_fpath +
' is empty or contains incorrect sequences (header-only or with non-ACGTN characters)!',
exit_with_code=1)
for excluded in excluded_ref_fpaths:
ref_fpaths.remove(excluded)
if len(chromosomes_by_refs) > 0:
logger.main_info(' All references were combined in ' + qconfig.combined_ref_name)
else:
logger.warning('All references were skipped!')
return corrected_ref_fpaths, combined_ref_fpath, chromosomes_by_refs, ref_fpaths
def correct_meta_references(ref_fpaths, corrected_dirpath):
corrected_ref_fpaths = []
combined_ref_fpath = os.path.join(corrected_dirpath, qconfig.combined_ref_name)
chromosomes_by_refs = {}
def _proceed_seq(seq_name, seq, ref_name, ref_fasta_ext, total_references, ref_fpath):
seq_fname = ref_name
seq_fname += ref_fasta_ext
if total_references > 1:
corr_seq_fpath = corrected_ref_fpaths[-1]
else:
corr_seq_fpath = qutils.unique_corrected_fpath(os.path.join(corrected_dirpath, seq_fname))
corrected_ref_fpaths.append(corr_seq_fpath)
corr_seq_name = qutils.name_from_fpath(corr_seq_fpath) + '_' + seq_name
if not qconfig.no_check:
corr_seq = correct_seq(seq, ref_fpath)
if not corr_seq:
return None, None
fastaparser.write_fasta(corr_seq_fpath, [(corr_seq_name, seq)], 'a')
fastaparser.write_fasta(combined_ref_fpath, [(corr_seq_name, seq)], 'a')
contigs_analyzer.ref_labels_by_chromosomes[corr_seq_name] = qutils.name_from_fpath(corr_seq_fpath)
chromosomes_by_refs[ref_name].append((corr_seq_name, len(seq)))
return corr_seq_name, corr_seq_fpath
ref_fnames = [os.path.basename(ref_fpath) for ref_fpath in ref_fpaths]
ref_names = []
for ref_fname in ref_fnames:
ref_name, ref_fasta_ext = qutils.splitext_for_fasta_file(ref_fname)
ref_names.append(ref_name)
dupl_ref_names = [ref_name for ref_name in ref_names if ref_names.count(ref_name) > 1]
for ref_fpath in ref_fpaths:
total_references = 0
ref_fname = os.path.basename(ref_fpath)
ref_name, ref_fasta_ext = qutils.splitext_for_fasta_file(ref_fname)
if ref_name in dupl_ref_names:
ref_name = qutils.get_label_from_par_dir_and_fname(ref_fpath)
chromosomes_by_refs[ref_name] = []
used_seq_names = defaultdict(int)
corr_seq_fpath = None
for i, (seq_name, seq) in enumerate(fastaparser.read_fasta(ref_fpath)):
total_references += 1
seq_name = correct_name(seq_name, qutils.MAX_CONTIG_NAME - len(ref_name) - 1)
uniq_seq_name = get_uniq_name(seq_name, used_seq_names)
used_seq_names[seq_name] += 1
corr_seq_name, corr_seq_fpath = _proceed_seq(uniq_seq_name, seq, ref_name, ref_fasta_ext, total_references, ref_fpath)
if not corr_seq_name:
break
if corr_seq_fpath:
logger.main_info(' ' + ref_fpath + ' ==> ' + qutils.name_from_fpath(corr_seq_fpath) + '')
logger.main_info(' All references combined in ' + qconfig.combined_ref_name)
return corrected_ref_fpaths, combined_ref_fpath, chromosomes_by_refs, ref_fpaths