sequences_process.py 文件源码

def __init__(self, logger, sequences, reference, dateFormat):
        super(sequence_set, self).__init__()
        self.log = logger

        # load sequences from the (parsed) JSON - don't forget to sort out dates
        self.seqs = {}
        for name, data in sequences.iteritems():
            self.seqs[name] = SeqRecord(Seq(data["seq"], generic_dna),
                   id=name, name=name, description=name)
            self.seqs[name].attributes = data["attributes"]
            # tidy up dates
            date_struc = parse_date(self.seqs[name].attributes["raw_date"], dateFormat)
            self.seqs[name].attributes["num_date"] = date_struc[1]
            self.seqs[name].attributes["date"] = date_struc[2]

        # if the reference is to be analysed it'll already be in the (filtered & subsampled)
        # sequences, so no need to add it here, and no need to care about attributes etc
        # we do, however, need it for alignment
        self.reference_in_dataset = reference["included"]
        name = reference["strain"]
        self.reference_seq = SeqRecord(Seq(reference["seq"], generic_dna),
               id=name, name=name, description=name)
        if "genes" in reference and len(reference["genes"]):
            self.proteins = {k:FeatureLocation(start=v["start"], end=v["end"], strand=v["strand"]) for k, v in reference["genes"].iteritems()}
        else:
            self.proteins = None

        # other things:
        self.run_dir = '_'.join(['temp', time.strftime('%Y%m%d-%H%M%S',time.gmtime()), str(random.randint(0,1000000))])
        self.nthreads = 2 # should load from config file