def tt_from_file(self, infile, root='best', nodefile=None):
self.is_timetree=False
self.logger('Reading tree from file '+infile,2)
dates = {seq.id:seq.attributes['num_date']
for seq in self.aln if 'date' in seq.attributes}
self.tt = TreeTime(dates=dates, tree=infile, gtr='Jukes-Cantor',
aln = self.aln, verbose=self.verbose, fill_overhangs=True)
if root:
self.tt.reroot(root=root)
self.tree = self.tt.tree
for node in self.tree.find_clades():
if node.is_terminal() and node.name in self.sequence_lookup:
seq = self.sequence_lookup[node.name]
node.attr = seq.attributes
try:
node.attr['date'] = node.attr['date'].strftime('%Y-%m-%d')
except:
pass
else:
node.attr = {}
if nodefile is not None:
self.logger('reading node properties from file: '+nodefile,2)
with myopen(nodefile, 'r') as infile:
from cPickle import load
node_props = load(infile)
for n in self.tree.find_clades():
if n.name in node_props:
for attr in node_props[n.name]:
n.__setattr__(attr, node_props[n.name][attr])
else:
self.logger("No node properties found for "+n.name,2)
def __init__(self, fname=None, reference_seq=None, **kwarks):
super(sequence_set, self).__init__()
self.kwarks = kwarks
self.nthreads = 2
if fname is not None and os.path.isfile(fname):
with myopen(fname) as seq_file:
self.all_seqs = {
x.name: x
for x in SeqIO.parse(seq_file, 'fasta')
}
elif 'virus' in kwarks:
self.from_vdb(kwarks['virus'])
else:
print('no input sequences found -- empty sequence set')
return
if 'run_dir' not in kwarks:
import random
self.run_dir = '_'.join([
'temp',
time.strftime('%Y%m%d-%H%M%S', time.gmtime()),
str(random.randint(0, 1000000))
])
else:
self.run_dir = kwarks['run_dir']
if reference_seq is not None:
if type(reference_seq) is str and reference_seq in self.all_seqs:
self.reference_seq = self.all_seqs[reference_seq]
else:
self.reference_seq = reference_seq
else:
self.reference_seq = None
def __init__(self, fname, reference=None, **kwarks):
super(sequence_set, self).__init__()
self.nthreads = 2
if os.path.isfile(fname):
with myopen(fname) as seq_file:
self.raw_seqs = {
fix_names(x.description): x
for x in SeqIO.parse(seq_file, 'fasta')
}
for x in self.raw_seqs.values():
x.id = fix_names(x.id)
x.name = fix_names(x.id)
x.description = fix_names(x.description)
if 'run_dir' not in kwarks:
import random
self.run_dir = '_'.join([
'temp',
time.strftime('%Y%m%d-%H%M%S', time.gmtime()),
str(random.randint(0, 1000000))
])
else:
self.run_dir = kwarks['run_dir']
if reference is not None:
if type(reference) is str and fix_names(
reference) in self.raw_seqs:
self.reference = self.raw_seqs[fix_names(reference)]
else:
self.reference = reference
else:
self.reference = None
def __init__(self, fname=None, reference= None, **kwarks):
super(sequence_set, self).__init__()
self.kwarks = kwarks
self.nthreads = 2
if fname is not None and os.path.isfile(fname):
with myopen(fname) as seq_file:
self.raw_seqs = {fix_names(x.description):x for x in SeqIO.parse(seq_file, 'fasta')}
for x in self.raw_seqs.values():
x.id = fix_names(x.id)
x.name = fix_names(x.id)
x.description = fix_names(x.description)
elif 'virus' in kwarks:
self.from_vdb(kwarks['virus'])
if 'run_dir' not in kwarks:
import random
self.run_dir = '_'.join(['temp', time.strftime('%Y%m%d-%H%M%S',time.gmtime()), str(random.randint(0,1000000))])
else:
self.run_dir = kwarks['run_dir']
if reference is not None:
if type(reference) is str and fix_names(reference) in self.raw_seqs:
self.reference = self.raw_seqs[fix_names(reference)]
else:
self.reference = reference
else: self.reference=None
def load_mfa(self, path):
try:
with myopen(path) as seq_file:
self.seqs = {x.name:x for x in SeqIO.parse(seq_file, 'fasta')}
except Exception as e:
self.log.fatal("Error loading sequences from {}. Error: {}".format(path, e))
self.nstart = len(self.seqs)
self.log.notify("Loaded {} sequences from {}".format(self.nstart, path))
def dump(self, treefile, nodefile):
from Bio import Phylo
Phylo.write(self.tree, treefile, 'newick')
node_props = {}
for node in self.tree.find_clades():
node_props[node.name] = {attr:node.__getattribute__(attr) for attr in self.dump_attr if hasattr(node, attr)}
with myopen(nodefile, 'w') as nfile:
from cPickle import dump
dump(node_props, nfile)
