def cluster(self, input_files, args, work_dir, threads=1):
J = lambda p: os.path.join(work_dir, p)
bin_prefix = J('METABAT_')
log_path = J('logs.txt')
cmd_line = [self.program_name,
'-i', input_files.contigs_fasta,
'-a', input_files.contig_coverages,
'-o', bin_prefix,
'--cvExt',
'-l',
*utils.serialize_args(args)]
self.progress.new(self.program_name)
self.progress.update('Running using %d threads...' % threads)
utils.run_command(cmd_line, log_path)
self.progress.end()
output_file_paths = glob.glob(J(bin_prefix + '*'))
if not len(output_file_paths):
raise ConfigError("Some critical output files are missing. Please take a look at the\
log file: %s" % (log_path))
clusters = {}
bin_count = 0
for bin_file in output_file_paths:
bin_count += 1
with open(bin_file, 'r') as f:
pretty_bin_name = os.path.basename(bin_file).replace('.', '_')
clusters[pretty_bin_name] = list(map(str.strip, f.readlines()))
return clusters
def cluster(self,
input_files,
args,
work_dir,
threads=1,
log_file_path=None):
J = lambda p: os.path.join(work_dir, p)
if not log_file_path:
log_file_path = J('logs.txt')
translation = {
'preference': 'p',
'maxiter': 'm',
'conviter': 'v',
'damp': 'd',
'contigsize': 'x'
}
cmd_line = [
self.program_name, '-c', input_files.contig_coverages_log_norm,
'-f',
os.path.dirname(input_files.contigs_fasta), '-l',
os.path.basename(input_files.contigs_fasta), '-o', work_dir,
*utils.serialize_args(
args, single_dash=True, translate=translation)
]
self.progress.new(self.program_name)
self.progress.update('Running using %d threads...' % threads)
utils.run_command(cmd_line, log_file_path)
self.progress.end()
output_file_paths = glob.glob(J('*.fna'))
if not len(output_file_paths):
raise ConfigError(
"Some critical output files are missing. Please take a look at the "
"log file: %s" % (log_file_path))
clusters = {}
bin_count = 0
for bin_file in output_file_paths:
bin_count += 1
with open(bin_file, 'r') as f:
pretty_bin_name = os.path.basename(bin_file)
pretty_bin_name = pretty_bin_name.replace('sequence_', '')
pretty_bin_name = pretty_bin_name.replace('.fna', '')
pretty_bin_name = pretty_bin_name.replace('-', '_')
clusters[pretty_bin_name] = [
line.strip().replace('>', '') for line in f
if line.startswith('>')
]
return clusters
def cluster(self,
input_files,
args,
work_dir,
threads=1,
log_file_path=None):
J = lambda p: os.path.join(work_dir, p)
output_file_prefix = J('MAXBIN_')
if not log_file_path:
log_file_path = J('logs.txt')
cmd_line = [
self.program_name, '-contig', input_files.contigs_fasta, '-abund',
input_files.contig_coverages, '-out', output_file_prefix,
'-thread',
str(threads),
*utils.serialize_args(args, single_dash=True, use_underscore=True)
]
self.progress.new(self.program_name)
self.progress.update('Running using %d threads...' % threads)
utils.run_command(cmd_line, log_file_path)
self.progress.end()
output_file_paths = glob.glob(J(output_file_prefix + '*.fasta'))
if not len(output_file_paths):
raise ConfigError(
"Some critical output files are missing. Please take a look at the "
"log file: %s" % (log_file_path))
clusters = {}
bin_count = 0
for bin_file in output_file_paths:
bin_count += 1
with open(bin_file, 'r') as f:
bin_name = os.path.basename(bin_file).replace('.fasta', '')
bin_name = bin_name.replace('.', '_')
clusters[bin_name] = []
for line in f.readlines():
if line.startswith('>'):
clusters[bin_name].append(line[1:].strip())
return clusters
def cluster(self, input_files, args, work_dir, threads=1, log_file_path=None):
J = lambda p: os.path.join(work_dir, p)
if not log_file_path:
log_file_path = J('logs.txt')
cmd_line = [self.program_name,
'--coverage_file', input_files.contig_coverages,
'--composition_file', input_files.contigs_fasta,
'--basename', work_dir,
'--threads', threads,
*utils.serialize_args(args, use_underscore=True)]
self.progress.new(self.program_name)
self.progress.update('Running using %d threads...' % threads)
utils.run_command(cmd_line, log_file_path)
self.progress.end()
clusters = {}
threshold = args.length_threshold or '1000'
output_file_name = 'clustering_gt%s.csv' % threshold
output_file_path = J(output_file_name)
if not os.path.exists(output_file_path):
raise ConfigError("One of the critical output files is missing ('%s'). Please take a look at the "
"log file: %s" % (output_file_name, log_file_path))
with open(output_file_path, 'r') as f:
lines = f.readlines()[1:]
for entry in lines:
contig, bin_name = map(str.strip, entry.split(','))
pretty_bin_name = 'Bin_' + bin_name
if pretty_bin_name not in clusters:
clusters[pretty_bin_name] = []
clusters[pretty_bin_name].append(contig)
return clusters
def cluster(self, input_files, args, work_dir, threads=1):
J = lambda p: os.path.join(work_dir, p)
cwd_backup = os.getcwd()
os.chdir(work_dir)
log_path = J('logs.txt')
c = ccollections.Collections(r=run, p=progress)
c.populate_collections_dict(input_files.profile_db)
source_collections = set(
map(str.strip, args.source_collections.split(',')))
missing_collections = source_collections - set(
c.collections_dict.keys())
if len(missing_collections):
raise ConfigError(
"Some of the collections you wanted are missing in the database. "
"Here is the list of missing collections: %s" %
(", ".join(missing_collections)))
c_names = []
c_files = []
for collection_name in source_collections:
prefix = J(collection_name)
c_names.append(collection_name)
c_files.append(prefix + '.txt')
c.export_collection(collection_name,
output_file_prefix=prefix,
include_unbinned=False)
cmd_line = [
self.program_name, '-c', input_files.splits_fasta, '-i',
','.join(c_files), '-l', ','.join(c_names), '-o',
J('OUTPUT'), '--threads',
str(threads), *utils.serialize_args(
args, use_underscore=True, skip_keys=['source_collections'])
]
self.progress.new(self.program_name)
self.progress.update('Running using %d threads...' % threads)
utils.run_command(cmd_line, log_path)
self.progress.end()
output_file_name = 'OUTPUT_DASTool_scaffolds2bin.txt'
output_file_path = J(output_file_name)
if not os.path.exists(output_file_path):
raise ConfigError(
"One of the critical output files is missing ('%s'). Please take a look at the "
"log file: %s" % (output_file_name, log_path))
clusters = {}
with open(output_file_path, 'r') as f:
lines = f.readlines()
for entry in lines:
contig, bin_name = map(str.strip, entry.split())
pretty_bin_name = 'Bin_' + bin_name.replace('.', '_')
if pretty_bin_name not in clusters:
clusters[pretty_bin_name] = []
clusters[pretty_bin_name].append(contig)
# restore cwd
os.chdir(cwd_backup)
return clusters