def summarise_bins(args):
message = '# CAT v{0}.'.format(about.__version__)
shared.give_user_feedback(message,
args.log_file,
args.quiet,
show_time=False)
errors = []
errors.append(
check.check_input_file(args.input_file, args.log_file, args.quiet))
if not args.force:
errors.append(
check.check_output_file(args.output_file, args.log_file,
args.quiet))
if True in errors:
sys.exit(1)
message = 'Summarising...'
shared.give_user_feedback(message, args.log_file, args.quiet)
with open(args.input_file, 'r') as f1:
for line in f1:
if line.startswith('#'):
line = line.split('\t')
if line[0] != '# bin':
message = '{0} is not a BAT classification file.'.format(
args.input_file)
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
if line[0] == '# contig':
message = (
'{0} appears to be a CAT classification file. '
'If you want to summarise contig '
'classifications, supply a contigs fasta with '
'argument [-c / --contigs_fasta].'.format(
args.input_file))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
try:
superkingdom_index = line.index('superkingdom')
except:
message = (
'official ranks not found in header of {0}. Make '
'sure that the BAT classification file is named '
'with official ranks with \'CAT add_names '
'--only_official\'.'.format(args.input_file))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
break
else:
message = 'input file does not have a recognisable header.'
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
n_bins = {}
n_bins['no taxid assigned'] = 0
official_ranks = [
'superkingdom', 'phylum', 'class', 'order', 'family', 'genus',
'species'
]
for rank in official_ranks:
n_bins[rank] = {}
n = 0
bin_trace = set()
doubles = set()
with open(args.input_file, 'r') as f1:
for line in f1:
line = line.rstrip()
if line.startswith('#'):
continue
n += 1
line = line.split('\t')
bin_ = line[0]
if bin_ in bin_trace:
doubles.add(bin_)
bin_trace.add(bin_)
if line[1] == 'no taxid assigned':
n_bins['no taxid assigned'] += 1
continue
for (i, classification) in enumerate(line[superkingdom_index:]):
classification = classification.rsplit(': ', 1)[0].rstrip('*')
rank = official_ranks[i]
if classification not in n_bins[rank]:
n_bins[rank][classification] = 0
n_bins[rank][classification] += 1
if len(doubles) != 0:
message = ('some bins have multiple classifications. CAT summarise '
'currently does not allow for this. Bins with multiple '
'classifications: {0}.'.format(', '.join(list(doubles))))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
n_classified_bins = n - n_bins['no taxid assigned']
with open(args.output_file, 'w') as outf1:
outf1.write('# total number of bins is {0:,d}, of which {1:,d} '
'({2:.2f}%) have taxonomy assigned.\n'.format(
n, n_classified_bins, n_classified_bins / n * 100))
outf1.write('#\n')
outf1.write('# rank\tclade\tnumber of bins\n')
for rank in official_ranks:
for clade in sorted(n_bins[rank],
key=lambda x: n_bins[rank][x],
reverse=True):
outf1.write('{0}\t{1}\t{2}\n'.format(rank, clade,
n_bins[rank][clade]))
message = '{0} is created!'.format(args.output_file)
shared.give_user_feedback(message, args.log_file, args.quiet)
return
def summarise_contigs(args):
message = '# CAT v{0}.'.format(about.__version__)
shared.give_user_feedback(message,
args.log_file,
args.quiet,
show_time=False)
errors = []
errors.append(
check.check_input_file(args.input_file, args.log_file, args.quiet))
if not args.force:
errors.append(
check.check_output_file(args.output_file, args.log_file,
args.quiet))
if True in errors:
sys.exit(1)
contig2length = import_contig_lengths(args.contigs_fasta, args.log_file,
args.quiet)
message = 'Summarising...'
shared.give_user_feedback(message, args.log_file, args.quiet)
with open(args.input_file, 'r') as f1:
for line in f1:
if line.startswith('#'):
line = line.split('\t')
if line[0] != '# contig':
message = '{0} is not a CAT classification file.'.format(
args.input_file)
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
if line[0] == '# bin':
message = (
'{0} appears to be a BAT classification file. '
'If you want to summarise bin '
'classifications, simply don\'t supply a '
'contigs fasta and everything should be fine.'
''.format(args.input_file))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
try:
superkingdom_index = line.index('superkingdom')
except:
message = (
'official ranks not found in header of {0}. Make '
'sure that the CAT classification file is named '
'with official ranks with \'CAT add_names '
'--only_official\'.'.format(args.input_file))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
break
else:
message = 'input file does not have a recognisable header.'
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
length = {}
length['no taxid assigned'] = []
ORFs = {}
official_ranks = [
'superkingdom', 'phylum', 'class', 'order', 'family', 'genus',
'species'
]
for rank in official_ranks:
length[rank] = {}
ORFs[rank] = {}
n = 0
contig_trace = set()
doubles = set()
with open(args.input_file, 'r') as f1:
for line in f1:
line = line.rstrip()
if line.startswith('#'):
continue
n += 1
line = line.split('\t')
contig = line[0]
if contig in contig_trace:
doubles.add(contig)
contig_trace.add(contig)
if contig not in contig2length:
message = (
'contig {0} in CAT classification file is not found '
'in supplied contigs fasta file. Are you sure the CAT '
'classification file is based on the contigs fasta?'
''.format(contig))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
if line[1] == 'no taxid assigned':
length['no taxid assigned'].append(contig2length[contig])
continue
for (i, classification) in enumerate(line[superkingdom_index:]):
classification = classification.rsplit(': ', 1)[0].rstrip('*')
rank = official_ranks[i]
if classification not in length[rank]:
length[rank][classification] = []
ORFs[rank][classification] = []
length[rank][classification].append(contig2length[contig])
# NOTE that the total number of ORFs on a contig is reproted,
# not only the number of ORFs a classification is based on.
ORFs_on_contig = int(line[2].split('/')[1].split(' ')[0])
ORFs[rank][classification].append(ORFs_on_contig)
if len(doubles) != 0:
message = ('some contigs have multiple classifications. CAT summarise '
'currently does not allow for this. Contigs with multiple '
'classifications: {0}.'.format(', '.join(list(doubles))))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
if n != len(contig2length):
message = ('the number of classified contigs is not the same as the '
'number of contigs in contigs fasta. Are you sure the CAT '
'classification file is based on the contigs fasta?')
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
with open(args.output_file, 'w') as outf1:
n_contigs = len(contig2length)
total_length = sum(contig2length.values())
n_classified_contigs = n_contigs - len(length['no taxid assigned'])
total_classified_length = total_length - sum(
length['no taxid assigned'])
outf1.write('# total number of contigs in {0} is {1:,d} representing '
'{2:,d} positions.\n'.format(args.contigs_fasta, n_contigs,
total_length))
outf1.write(
'# {0:,d} contigs have taxonomy assigned ({1:.2f}%) '
'representing {2:,d} positions ({3:.2f}%) in {4}.\n'.format(
n_classified_contigs, n_classified_contigs / n_contigs * 100,
total_classified_length,
total_classified_length / total_length * 100, args.input_file))
outf1.write('#\n')
outf1.write('# rank\t'
'clade\t'
'number of contigs\t'
'number of ORFs\t'
'number of positions\n')
for rank in official_ranks:
for clade in sorted(length[rank],
key=lambda x: sum(length[rank][x]),
reverse=True):
outf1.write('{0}\t{1}\t{2}\t{3}\t{4}\n'.format(
rank, clade, len(length[rank][clade]),
sum(ORFs[rank][clade]), sum(length[rank][clade])))
message = '{0} is created!'.format(args.output_file)
shared.give_user_feedback(message, args.log_file, args.quiet)
return
def run():
args = parse_arguments()
message = '# CAT v{0}.'.format(about.__version__)
shared.give_user_feedback(message,
args.log_file,
args.quiet,
show_time=False)
errors = []
errors.append(
check.check_input_file(args.input_file, args.log_file, args.quiet))
if not args.force:
errors.append(
check.check_output_file(args.output_file, args.log_file,
args.quiet))
if True in errors:
sys.exit(1)
(taxid2parent, taxid2rank) = tax.import_nodes(args.nodes_dmp,
args.log_file, args.quiet)
taxid2name = tax.import_names(args.names_dmp, args.log_file, args.quiet)
message = 'Appending names...'
shared.give_user_feedback(message, args.log_file, args.quiet)
with open(args.input_file, 'r') as f1:
for line in f1:
if line.startswith('#'):
line = line.rstrip().split('\t')
if 'lineage' in line:
lineage_index = line.index('lineage')
else:
message = ('{0} is not a supported classification file.'
''.format(input_file))
shared.give_user_feedback(message,
args.log_file,
args.quiet,
error=True)
sys.exit(1)
try:
scores_index = line.index('lineage scores')
except:
scores_index = None
full_length = len(line)
break
else:
message = ('{0} is not a supported classification file.'.format(
args.input_file))
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
with open(args.input_file, 'r') as f1, open(args.output_file,
'w') as outf1:
for line in f1:
line = line.rstrip()
if line.startswith('#'):
if args.only_official:
outf1.write('{0}\tsuperkingdom\tphylum\tclass\torder\t'
'family\tgenus\tspecies\n'.format(line))
else:
outf1.write('{0}\tfull lineage names\n'.format(line))
continue
line = line.split('\t')
if len(line) != full_length:
# Entry does not have a full annotation.
outf1.write('{0}\n'.format('\t'.join(line)))
continue
if (line[1].startswith('no taxid found')
or line[2].startswith('no taxid found')):
# ORF has database hits but the accession number is not found
# in the taxonomy files.
outf1.write('{0}\n'.format('\t'.join(line)))
continue
lineage = line[lineage_index].split(';')
if scores_index is not None and not args.exclude_scores:
scores = line[scores_index].split(';')
else:
scores = None
if args.only_official:
names = tax.convert_to_official_names(lineage, taxid2rank,
taxid2name, scores)
else:
names = tax.convert_to_names(lineage, taxid2rank, taxid2name,
scores)
outf1.write('{0}\t{1}\n'.format('\t'.join(line), '\t'.join(names)))
message = 'Names written to {0}!'.format(args.output_file)
shared.give_user_feedback(message, args.log_file, args.quiet)
return
def summarise_bins(input_file, output_file, force, quiet):
# Currently summarise does not a allow for a log file.
log_file = None
message = '# CAT v{0}.'.format(about.__version__)
shared.give_user_feedback(message, log_file, quiet, show_time=False)
errors = []
errors.append(check.check_input_file(input_file, log_file, quiet))
if not force:
errors.append(check.check_output_file(output_file, log_file, quiet))
if True in errors:
sys.exit(1)
message = 'Summarising...'
shared.give_user_feedback(message, log_file, quiet)
with shared.open_maybe_gzip(input_file, 'rt') as f1:
for line in f1:
if line.startswith('#'):
line = line.split('\t')
if line[0] != '# bin':
message = ('ERROR: {0} is not a BAT classification file.'
''.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
if line[0] == '# contig':
message = ('ERROR: {0} appears to be a CAT '
'classification file. If you want to '
'summarise contig classifications, please '
'supply a contigs fasta.'
''.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
sys.exit(1)
try:
superkingdom_index = line.index('superkingdom')
except:
message = ('ERROR: official ranks not found in header of '
'{0}. Make sure that the BAT classification '
'file is named with official ranks with \'CAT '
'add_names --only_official\'.'
''.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
sys.exit(1)
break
else:
message = 'ERROR: input file does not have a recognisable header.'
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
number_of_bins = {}
number_of_bins['unclassified'] = 0
official_ranks = [
'superkingdom', 'phylum', 'class', 'order', 'family', 'genus',
'species'
]
for rank in official_ranks:
number_of_bins[rank] = {}
n = 0
bin_trace = set()
doubles = set()
with open(input_file, 'r') as f1:
for line in f1:
line = line.rstrip()
if line.startswith('#'):
continue
n += 1
line = line.split('\t')
bin_ = line[0]
if bin_ in bin_trace:
doubles.add(bin_)
bin_trace.add(bin_)
if line[1] == 'unclassified':
number_of_bins['unclassified'] += 1
continue
for (i, classification) in enumerate(line[superkingdom_index:]):
classification = classification.rsplit(': ', 1)[0].rstrip('*')
rank = official_ranks[i]
if classification not in number_of_bins[rank]:
number_of_bins[rank][classification] = 0
number_of_bins[rank][classification] += 1
if len(doubles) != 0:
message = ('ERROR: some bins have multiple classifications. CAT '
'summarise currently does not allow for this. Bins with '
'multiple classifications: {0}.'
''.format(', '.join(list(doubles))))
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
number_of_classified_bins = n - number_of_bins['unclassified']
with shared.open_maybe_gzip(output_file, 'wt') as outf1:
outf1.write('# total number of bins is {0}, of which {1} ({2:.2f}%) '
'are classified.\n'
''.format(n, number_of_classified_bins,
number_of_classified_bins / n * 100))
outf1.write('#\n')
outf1.write('# rank\tclade\tnumber of bins\n')
for rank in official_ranks:
for clade in sorted(number_of_bins[rank],
key=lambda x: number_of_bins[rank][x],
reverse=True):
outf1.write('{0}\t{1}\t{2}\n'
''.format(rank, clade,
number_of_bins[rank][clade]))
message = '{0} is created!'.format(output_file)
shared.give_user_feedback(message, log_file, quiet)
def add_names(args):
(input_file, output_file, taxonomy_folder, only_official, exclude_scores,
force, quiet) = check.convert_arguments(args)
# Currently add_names does not allow for a log file.
log_file = None
message = '# CAT v{0}.'.format(about.__version__)
shared.give_user_feedback(message, log_file, quiet, show_time=False)
errors = []
errors.append(check.check_input_file(input_file, log_file, quiet))
if not force:
errors.append(check.check_output_file(output_file, log_file, quiet))
if True in errors:
sys.exit(1)
(nodes_dmp, names_dmp, prot_accession2taxid_file
) = check.inspect_taxonomy_folder(taxonomy_folder)
(taxid2parent, taxid2rank) = tax.import_nodes(nodes_dmp, log_file, quiet)
taxid2name = tax.import_names(names_dmp, log_file, quiet)
message = 'Appending names...'
shared.give_user_feedback(message, log_file, quiet)
with shared.open_maybe_gzip(input_file, 'rt') as f1:
for line in f1:
if line.startswith('#'):
line = line.rstrip().split('\t')
try:
lineage_index = line.index('lineage')
except:
message = ('ERROR: {0} is not a supported classification '
'file.'.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
sys.exit(1)
try:
scores_index = line.index('lineage scores')
except:
scores_index = None
full_length = len(line)
break
else:
message = ('ERROR: {0} is not a supported classification file.'
''.format(input_file))
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
with shared.open_maybe_gzip(input_file,
'rt') as f1, shared.open_maybe_gzip(
output_file, 'wt') as outf1:
for line in f1:
line = line.rstrip()
if line.startswith('#'):
if only_official:
outf1.write('{0}\tsuperkingdom\tphylum\tclass\torder\t'
'family\tgenus\tspecies\n'.format(line))
else:
outf1.write('{0}\tfull lineage names\n'.format(line))
continue
line = line.split('\t')
if len(line) != full_length:
# Entry does not have a full annotation.
outf1.write('{0}\n'.format('\t'.join(line)))
continue
if (line[1].startswith('no taxid found')
or line[2].startswith('no taxid found')):
# ORF has database hits but the accession number is not found
# in the taxonomy files.
outf1.write('{0}\n'.format('\t'.join(line)))
continue
lineage = line[lineage_index].split(';')
if scores_index and not exclude_scores:
scores = line[scores_index].split(';')
else:
scores = None
if only_official:
names = tax.convert_to_official_names(lineage, taxid2rank,
taxid2name, scores)
else:
names = tax.convert_to_names(lineage, taxid2rank, taxid2name,
scores)
outf1.write('{0}\t{1}\n'.format('\t'.join(line), '\t'.join(names)))
message = 'Names written to {0}!'.format(output_file)
shared.give_user_feedback(message, log_file, quiet)
def summarise_contigs(input_file, output_file, contigs_fasta, force, quiet):
# Currently summarise does not a allow for a log file.
log_file = None
message = '# CAT v{0}.'.format(about.__version__)
shared.give_user_feedback(message, log_file, quiet, show_time=False)
errors = []
errors.append(check.check_input_file(input_file, log_file, quiet))
if not force:
errors.append(check.check_output_file(output_file, log_file, quiet))
if True in errors:
sys.exit(1)
contig2length = import_contig_lengths(contigs_fasta, log_file, quiet)
message = 'Summarising...'
shared.give_user_feedback(message, log_file, quiet)
with open(input_file, 'r') as f1:
for line in f1:
if line.startswith('#'):
line = line.split('\t')
if line[0] != '# contig':
message = ('ERROR: {0} is not a CAT classification file.'
''.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
if line[0] == '# bin':
message = ('ERROR: {0} appears to be a BAT '
'classification file. If you want to '
'summarise bin classifications, just '
'don\'t supply a contigs fasta and '
'everything should be fine!'
''.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
sys.exit(1)
try:
superkingdom_index = line.index('superkingdom')
except:
message = ('ERROR: official ranks not found in header of '
'{0}. Make sure that the CAT classification '
'file is named with official ranks with \'CAT '
'add_names --only_official\'.'
''.format(input_file))
shared.give_user_feedback(message,
log_file,
quiet,
error=True)
sys.exit(1)
break
else:
message = 'ERROR: input file does not have a recognisable header.'
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
length = {}
length['unclassified'] = []
ORFs = {}
official_ranks = ['superkingdom', 'phylum', 'class', 'order', 'family',
'genus', 'species']
for rank in official_ranks:
length[rank] = {}
ORFs[rank] = {}
n = 0
contig_trace = set()
doubles = set()
with open(input_file, 'r') as f1:
for line in f1:
line = line.rstrip()
if line.startswith('#'):
continue
n += 1
line = line.split('\t')
contig = line[0]
if contig in contig_trace:
doubles.add(contig)
contig_trace.add(contig)
if contig not in contig2length:
message = ('ERROR: contig {0} in CAT classification file is '
'not found in supplied contigs fasta file. Are you '
'sure the CAT classification file is based on the '
'contigs fasta?'.format(contig))
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
if line[1].startswith('unclassified'):
length['unclassified'].append(contig2length[contig])
continue
for (i, classification) in enumerate(line[superkingdom_index:]):
classification = classification.rsplit(': ', 1)[0].rstrip('*')
rank = official_ranks[i]
if classification not in length[rank]:
length[rank][classification] = []
ORFs[rank][classification] = []
length[rank][classification].append(contig2length[contig])
ORFs[rank][classification].append(int(line[2]))
if len(doubles) != 0:
message = ('ERROR: some contigs have multiple classifications. CAT '
'summarise currently does not allow for this. Contigs with '
'multiple classifications: {0}.'
''.format(', '.join(list(doubles))))
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
if n != len(contig2length):
message = ('ERROR: the number of classified contigs is not the same '
'as the number of contigs in contigs fasta. Are you sure '
'the CAT classification file is based on the contigs '
'fasta?')
shared.give_user_feedback(message, log_file, quiet, error=True)
sys.exit(1)
with open(output_file, 'w') as outf1:
number_of_contigs = len(contig2length)
total_length = sum(contig2length.values())
number_of_classified_contigs = number_of_contigs - len(length['unclassified'])
total_classified_length = total_length - sum(length['unclassified'])
outf1.write('# total number of contigs in {0} is {1} representing {2} '
'positions.\n'
''.format(contigs_fasta,
number_of_contigs,
total_length))
outf1.write('# {0} contigs are classified ({1:.2f}%) representing {2} '
'positions ({3:.2f}%) in {4}.\n'
''.format(number_of_classified_contigs,
number_of_classified_contigs / number_of_contigs * 100,
total_classified_length,
total_classified_length / total_length * 100,
input_file))
outf1.write('#\n')
outf1.write('# rank\t'
'clade\t'
'number of contigs\t'
'number of ORFs\t'
'number of positions\n')
for rank in official_ranks:
for clade in sorted(length[rank],
key=lambda x: sum(length[rank][x]),
reverse=True):
outf1.write('{0}\t{1}\t{2}\t{3}\t{4}\n'
''.format(rank,
clade,
len(length[rank][clade]),
sum(ORFs[rank][clade]),
sum(length[rank][clade])))
message = '{0} is created!'.format(output_file)
shared.give_user_feedback(message, log_file, quiet)