def check_gzip(genome_path):
# One final check to see all downloaded files are unpacked
logger.debug('Starting function "check_gzip"')
for folder in os.listdir(genome_path):
f = find_file(os.path.join(genome_path, folder), ['.gz'], [])
if f:
unpack(f)
def prodigal_worker(genome, settings, path):
genome_path = os.path.join(path, genome)
sco_file = os.path.join(genome_path, genome + '.sco')
fasta_file = find_file(genome_path, ['fna', 'fasta'], ['genomic'])
fasta_file = os.path.join(genome_path, fasta_file)
out_genbank = os.path.join(genome_path, genome + '_prodigal.gbk')
if settings['override_prodigal'] or not os.path.isfile(sco_file):
# Always run if override_prodigal is set to do so, else only run if no sco file is found
run_prodigal_cmd(fasta_file, sco_file)
sr = SeqIO.parse(fasta_file, 'fasta')
features, gene_nr, gene_names = parse_prodigal(sco_file,
start_nr=1,
prefix=genome + '_')
all_seqs = []
scaf_names_count = {}
for record in sr:
renamed = False
scaffold_name = record.id
if scaffold_name not in scaf_names_count:
scaf_names_count[scaffold_name] = 0
scaf_names_count[scaffold_name] += 1
scaffold_features = features[scaffold_name]
record.features = scaffold_features
all_seqs.append(record)
if not os.path.isfile(out_genbank):
write_genbank_single(all_seqs, out_genbank)
return (genome, all_seqs, out_genbank, gene_names, scaf_names_count)
def process_from_fasta(genomes, path, override_prodigal=False):
# Run prodigal for each genome
# Parse the fasta via biopython
# Make a feature list based on prodigals output and expand the seqrecord with it
# Pass it onto regular genbank parsing
genome_seqs = {}
all_names = set()
gene_nr = 1
for genome in genomes:
genome_path = os.path.join(path, genome)
sco_file = os.path.join(genome_path, genome + '.sco')
fasta_file = find_file(genome_path, ['fna', 'fasta'], ['genomic'])
fasta_file = os.path.join(genome_path, fasta_file)
if override_prodigal or not os.path.isfile(sco_file):
# Always run if override_prodigal is set to do so, else only run if no sco file is found
run_prodigal_cmd(fasta_file, sco_file)
sr = SeqIO.parse(fasta_file, 'fasta')
features, gene_nr, names = parse_prodigal(sco_file, prev_nr=gene_nr)
all_names = fuse_dict_add(all_names, names)
all_seqs = []
for record in sr:
name = record.id
scaffold_features = features[name]
record.features = scaffold_features
all_seqs.append(record)
genome_seqs[genome] = all_seqs
# Get all scaffold names and gene names so that they can be checked for double names laters
scaffold_names = get_scaffold_names(genome_seqs)
# Write the files to a genbank file; also adds translations to the features
files_parsed = write_genbank(genome_seqs, path)
logger.debug('Prodigal parsing: renamed %i genes' % (gene_nr - 1))
return (genome_seqs, all_names, scaffold_names, files_parsed)
def find_downloaded_files(genome_path, prodigal, ext=False):
logger.debug('Starting function "find_downloaded_files"')
already_dl = []
for folder in os.listdir(genome_path):
if ext:
exts = [ext] + [ext.rpartition('.gz')[0]]
else:
if prodigal == 'never':
exts = ['.gbk', '.gbff', '.gbk.gz', '.gbff.gz']
elif prodigal == 'always':
exts = ['.fna', '.fna.gz']
else:
exts = [
'.gbk', '.gbk.gz', '.gbff', '.gbff.gz', '.fna', '.fna.gz'
]
f = find_file(os.path.join(genome_path, folder), exts, [])
if f:
already_dl.append(folder)
return already_dl