parser.add_argument("json", help="prepared JSON from augur")
parser.add_argument("--metadata", help="tab-delimited file to dump prepared JSON metadata to")
args = parser.parse_args()
# Setup the logger.
logger = logging.getLogger(__name__)
# Load the JSON data.
with open(args.json, "r") as fh:
data = json.load(fh)
# Prepare a sequence set.
sequences = sequence_set(
logger,
data["sequences"],
data["reference"],
data["info"]["date_format"]
)
# Write sequences to standard out.
output_sequences = sequences.seqs.values()
Bio.SeqIO.write(output_sequences, sys.stdout, "fasta")
# Prepare metadata if it has been requested.
if args.metadata:
metadata = [sequences.seqs[seq].attributes for seq in sequences.seqs]
metadata_df = pd.DataFrame(metadata)
metadata_df = metadata_df.rename(columns={"num_date": "prepared_num_date"})
metadata_df.to_csv(args.metadata, sep="\t", index=False)
def __init__(self, config):
""" check config file, make necessary directories, set up logger """
super(process, self).__init__()
self.config = combine_configs("process", config)
try:
assert (os.path.basename(os.getcwd()) == self.config["dir"])
except AssertionError:
print("Run this script from within the {} directory".format(
self.config["dir"]))
sys.exit(2)
for p in self.config["output"].values():
if not os.path.isdir(p):
os.makedirs(p)
self.log = logger(self.config["output"]["data"], False)
# parse the JSON into different data bits
try:
with open(self.config["in"], 'r') as fh:
data = json.load(fh)
except Exception as e:
self.log.fatal("Error loading JSON. Error: {}".format(e))
self.info = data["info"]
if "time_interval" in data["info"]:
self.info["time_interval"] = [
datetime.strptime(x, '%Y-%m-%d').date()
for x in data["info"]["time_interval"]
]
self.info["lineage"] = data["info"]["lineage"]
if 'leaves' in data:
self.tree_leaves = data['leaves']
try:
self.colors = data["colors"]
except KeyError:
self.log.notify("* colours have not been set")
self.colors = False
try:
self.lat_longs = data["lat_longs"]
except KeyError:
self.log.notify("* latitude & longitudes have not been set")
self.lat_longs = False
# backwards compatability - set up file_dumps (need to rewrite sometime)
# self.sequence_fname = self.input_data_path+'.fasta'
self.file_dumps = {}
self.output_path = os.path.join(self.config["output"]["data"],
self.info["prefix"])
self.file_dumps['seqs'] = self.output_path + '_sequences.pkl.gz'
self.file_dumps['tree'] = self.output_path + '_tree.newick'
self.file_dumps['nodes'] = self.output_path + '_nodes.pkl.gz'
if self.config["clean"] == True:
self.log.notify("Removing intermediate files for a clean build")
for f in glob.glob(self.output_path + "*"):
os.remove(f)
if "reference" in data:
self.seqs = sequence_set(self.log, data["sequences"],
data["reference"],
self.info["date_format"])
else:
self.log.fatal("No reference provided. Cannot continue.")
# self.seqs = sequence_set(self.log, data["sequences"], False, self.info["date_format"])
# backward compatability
self.reference_seq = self.seqs.reference_seq
self.proteins = self.seqs.proteins
for trait in self.info["traits_are_dates"]:
self.seqs.convert_trait_to_numerical_date(trait,
self.info["date_format"])
# Prepare titers if they are available.
if "titers" in data:
self.log.debug("Loaded %i titer measurements" %
len(data["titers"]))
# Convert titer dictionary indices from JSON-compatible strings back
# to tuples.
self.titers = {
eval(key): value
for key, value in data["titers"].iteritems()
}
## usefull flag to set (from pathogen run file) to disable restoring
self.try_to_restore = True