def __call__(self, seqs):
'It runs the actual annotations'
if not seqs:
return seqs
pep_fhand = NamedTemporaryFile()
dna_fhand = NamedTemporaryFile()
_run_estscan(seqs, pep_fhand.name, dna_fhand.name, self._usage_matrix)
# now we read the result files
estscan_result = _read_estcan_results(open(pep_fhand.name),
open(dna_fhand.name))
for seq in seqs:
seq_name = get_name(seq)
orfs = estscan_result.get(seq_name, {})
feats = []
for (start, end, strand), str_seqs in orfs.viewitems():
start -= 1
# end is fine -- end[
feat = SeqFeature(location=FeatureLocation(start, end, strand),
type='ORF',
qualifiers=str_seqs)
feats.append(feat)
if feats:
seq.object.features.extend(feats)
dna_fhand.close()
pep_fhand.close()
return seqs
def _annotate_polya(seq, min_len, max_cont_mismatches):
'It annotates the polyA with the EMBOSS trimest method'
str_seq = get_str_seq(seq)
polya = _detect_polya_tail(str_seq, THREE_PRIME, min_len,
max_cont_mismatches)
polyt = _detect_polya_tail(str_seq, FIVE_PRIME, min_len,
max_cont_mismatches)
a_len = polya[1] - polya[0] if polya else 0
t_len = polyt[1] - polyt[0] if polyt else 0
chosen_tail = None
if a_len > t_len:
chosen_tail = 'A'
elif t_len > a_len:
chosen_tail = 'T'
elif a_len and a_len == t_len:
if randint(0, 1):
chosen_tail = 'A'
else:
chosen_tail = 'T'
if chosen_tail:
strand = 1 if chosen_tail == 'A' else -1
start, end = polya if chosen_tail == 'A' else polyt
feat = SeqFeature(location=FeatureLocation(start, end, strand),
type='polyA_sequence')
# We're assuming that the seq has a SeqRecord in it
seq.object.features.append(feat)
def __call__(self, seqrecords):
'It does the work'
if not seqrecords:
return seqrecords
matcher = Blaster(seqrecords,
self.blastdb,
self._program,
self._dbtype,
filters=self._filters,
params=self._params,
remote=self._remote)
blasts = matcher.blasts
blastdb = os.path.basename(self.blastdb)
for seqrecord in seqrecords:
align_result = blasts.get(get_name(seqrecord), None)
if not align_result:
continue
match_counter = 0
for match in align_result['matches']:
subject = match['subject']['name']
match_counter += 1
for match_part in match['match_parts']:
if match_part['subject_end'] < match_part['subject_start']:
strand = -1
subject_start = match_part['subject_end']
subject_end = match_part['subject_start']
else:
strand = 1
subject_start = match_part['subject_start']
subject_end = match_part['subject_end']
query_start = match_part['query_start']
query_end = match_part['query_end']
qualifiers = {}
qualifiers['Target'] = {
'start': subject_start,
'end': subject_end,
'name': subject
}
qualifiers['score'] = match_part['scores']['expect']
qualifiers['identity'] = match_part['scores']['identity']
qualifiers['blastdb'] = blastdb
location = FeatureLocation(query_start, query_end, strand)
feature = SeqFeature(
location=location,
type='match_part',
qualifiers=qualifiers,
id='match{0:03d}'.format(match_counter))
seqrecord.object.features.append(feature)
return seqrecords