def main(args):
tables = []
container = bioio.multisequence()
n = 1
for infile in args.files:
mseqs = bioio.load(infile)
mseqs.sort(lambda x: x.label)
for s in mseqs:
tables.append((
n,
s.label,
s.attr.get('collection_date', ''),
s.attr.get('country', ''),
s.attr.get('isolate', ''),
s.definition,
))
container.append(bioio.biosequence('%04d' % n, s.seq.upper()))
n += 1
# write to output file
tabfile = open(args.tabfile, 'w')
tabfile.write('LABEL\tACCNO\tDATE\tCOUNTRY\tISOLATE\tDEFINITION\n')
tables.sort()
for r in tables:
tabfile.write('%04d\t%s\t%s\t%s\t%s\t%s\n' % r)
tabfile.close()
bioio.save(container, args.outfile)
def new_sequence(self):
n = bioio.biosequence('~new~', b'')
m = self.model()
s = m.selection()
if s is not None:
idx = max( s.indices() )
m.insert( idx, n )
else:
m.append(n)
m.signals().ContentUpdated.emit()
def retranslate(self, idx=None):
if not idx:
self.clear()
for idx in range( len( self._src_msa ) ):
self.append(
bioio.biosequence('', funcs.translated( self._src_msa[idx].seq,
start_pos = self._start_atg )) )
print(self[-1].seq)
else:
self[idx].seq = funcs.translated( self._src_msa[idx].seq,
start_pos = self._start_atg )
def make_consensus(self, thresholds = [ 0.5, 0.75, 0.90, 0.95, 0.99] ):
from seqpy.core.funcs.profiles import na_profile, aa_profile
m = self.model()
if isinstance(m.selection(), LabelSelection):
msa = m.selection().copy_to_msa()
else:
msa = m
if m.type() == bioio.PROTEIN:
profile = aa_profile(msa)
else:
profile = na_profile(msa)
for th in thresholds:
m.append(bioio.biosequence('CONS-%3.2f' % th, profile.consensus(th)))
m.signals().ContentUpdated.emit()
def dereplicate(mseq):
from seqpy.core.bioio import biosequence, multisequence
dedups = {}
for s in mseq:
if str(s.seq) in dedups:
dedups[str(s.seq)][1].append( s.label )
else:
dedups[str(s.seq)] = (s.seq, [ s.label ] )
dedupseqs = multisequence()
for (k, v) in dedups.items():
dedupseqs.append( biosequence( '#'.join( v[1] ), v[0] ) )
return dedupseqs
def main(args):
mseq = bioio.multisequence()
for infile in args.files:
trace = bioio.load(infile)
result = traceutils.trim(trace, args.winsize, args.qual_threshold)
if not result:
continue
bases, quals, upstream_trim, downstream_trim = result
seq = bioio.biosequence(infile, bases)
seq.add_attr('upstream_trim', str(upstream_trim))
seq.add_attr('downstream_trim', str(downstream_trim))
mseq.append(seq)
bioio.save(mseq, args.outfile)
def condensed(multiseq):
from seqpy.core.bioio import biosequence
cseqs = multiseq.clone()
positions = []
for x in range( multiseq.max_seqlen() ):
c = multiseq[0][x]
for s in multiseq[1:]:
if s[x] == 78 or s[x] == 110:
continue
if s[x] != c and abs(s[x] - c) != 32:
break
else:
continue
for y in range(len(multiseq)):
cseqs[y].append( multiseq[y][x] )
positions.append( x )
cseqs.add_control('position', biosequence('position', b','.join( ('%d' % x).encode('ASCII') for x in positions)))
return cseqs
def make_consensus_marker(self):
from seqpy.core.funcs.profiles import na_profile, aa_profile
m = self.model()
if isinstance(m.selection(), LabelSelection):
msa = m.selection().copy_to_msa()
else:
msa = m
if m.type() == bioio.PROTEIN:
profile = aa_profile(msa)
else:
profile = na_profile(msa)
cons_seq = bytearray()
cons = profile.mat
for i in range(0, len(cons)):
max_pos = cons[i].argmax()
if cons[i, max_pos] < 0.99:
cons_seq.append( 42 )
print(i)
else:
cons_seq.append( 32 )
m.append(bioio.biosequence('CONS-MARKER', cons_seq))
m.signals().ContentUpdated.emit()
def init_samples(self, line):
super().init_samples(line)
# create multisequence and populate with similar samples
for label in self.sample_labels:
self.mseq.append(bioio.biosequence(label=label))
def get_sequence(self):
from seqpy.core.bioio import biosequence
return biosequence(self.name(), self.edit_bases)