def alleleretriever(self):
"""
Retrieve the required alleles from a file of all alleles, and create organism-specific allele files
"""
logging.info('Retrieving alleles')
# Index all the records in the allele file
logging.info('Loading rMLST records')
recorddict = SeqIO.index(self.allelefile, 'fasta')
logging.info('Creating allele output files')
# Create the organism-specific files of alleles
for organism in sorted(self.alleledict):
# Make an object to store information for each strain
metadata = MetadataObject()
metadata.organism = organism
metadata.path = self.path
metadata.outpath = os.path.join(self.path, 'outputalleles',
organism, '')
# Delete and recreate the output path - as the files are appended to each time, they will be too large if
# this script is run more than once
try:
shutil.rmtree(metadata.outpath)
except OSError:
pass
make_path(metadata.outpath)
metadata.combined = os.path.join(metadata.outpath,
'gdcs_alleles.fasta')
metadata.allelefiles = list()
with open(metadata.combined, 'w') as combined:
for gene, alleles in sorted(self.alleledict[organism].items()):
# Open the file to append
allelefiles = os.path.join(metadata.outpath,
'{}.tfa'.format(gene))
metadata.allelefiles.append(allelefiles)
with open(allelefiles, 'a') as allelefile:
# Write each allele record to the file
for allele in sorted(alleles):
# Skip adding alleles that are no longer in the database
try:
SeqIO.write(
recorddict['{}_{}'.format(gene, allele)],
allelefile, 'fasta')
SeqIO.write(
recorddict['{}_{}'.format(gene, allele)],
combined, 'fasta')
except KeyError:
pass
# Add the populated metadata to the list
self.samples.append(metadata)
def rmlst(self):
"""
Get the most up-to-date profiles and alleles from pubmlst. Note that you will need the necessary access token
and secret for this to work
"""
printtime('Downloading rMLST database', self.start)
# Set the name of the file to be used to determine if the database download and set-up was successful
completefile = os.path.join(self.databasepath, 'rMLST', 'complete')
if not os.path.isfile(completefile):
# Create an object to send to the rMLST download script
args = MetadataObject()
# Add the path and start time attributes
args.path = self.databasepath
args.start = self.start
# Run the rMLST download
get_rmlst.Get(args)
# Create and populate the complete.txt file
with open(completefile, 'w') as complete:
complete.write('\n'.join(glob(os.path.join(self.databasepath, 'rMLST', '*'))))
def __init__(self, inputobject, extension='fasta', light=False):
# Create an object to mimic the command line arguments necessary for the script
args = MetadataObject()
args.path = inputobject.path
args.sequencepath = inputobject.path
args.databasepath = os.path.join(inputobject.reffilepath, 'clark')
make_path(args.databasepath)
args.clarkpath = os.path.dirname(which('CLARK'))
args.clarkpath += '/../opt/clark/'
args.cutoff = 0.005
args.database = 'bacteria'
args.rank = 'species'
args.filter = False
args.threads = inputobject.cpus
args.runmetadata = inputobject.runmetadata
args.clean_seqs = False
args.reffilepath = inputobject.reffilepath
args.runmetadata.extension = extension
args.light = light
# Run CLARK
CLARK(args, inputobject.commit, inputobject.starttime,
inputobject.homepath)
def mlst(self, genera={'Escherichia', 'Vibrio', 'Campylobacter', 'Listeria', 'Bacillus', 'Staphylococcus',
'Salmonella'}):
"""
Download the necessary up-to-date MLST profiles and alleles
"""
printtime('Downloading MLST databases', self.start)
for genus in genera:
# Create an object to pass to the get_mlst script
args = MetadataObject()
# Populate the object with the necessary attributes
args.species = genus
args.repository_url = 'http://pubmlst.org/data/dbases.xml'
args.force_scheme_name = False
args.path = os.path.join(self.databasepath, 'MLST', genus)
# Create the name of the file to be used to determine if the database download and setup was successful
completefile = os.path.join(args.path, 'complete')
# Only download the files if the download was not previously successful
if not os.path.isfile(completefile):
# Run the download
get_mlst.main(args)
# Create and populate the complete.txt file
with open(completefile, 'w') as complete:
complete.write('\n'.join(glob(os.path.join(args.path, '*'))))