def _asp(request):
try:
organism = str(request.POST['organism'])
sequence = str(request.POST['sequence'])
start = int(request.POST['start'])
stop = int(request.POST['stop'])
strand = str(request.POST['strand'])
feature = str(request.POST['feature'])
if strand == '-':
strand = '-'
else:
strand = '+'
modelfname = 'data/asp/%s.dat.bz2' % organism
# length limit to no smaller than 10 byte and no bigger than 2M
if len(sequence) <= 10 or len(sequence) > 2 * 1024 * 1024:
raise ValueError("fasta file incorrect")
if not os.path.isfile(modelfname):
raise ValueError(
"model should be one of Cress, Fish, Fly, Human, Worm")
if start < 80:
raise ValueError("--start value must be >=80")
if stop > 0 and start >= stop - 80:
raise ValueError("--stop value must be > start + 80")
if stop < 0 and stop > -80:
raise ValueError("--stop value must be <= - 80")
# shift the start and stop a bit
start -= 1
stop -= 1
sequence_buf = StringIO.StringIO()
sequence_buf.write(sequence)
sequence_buf.seek(0)
fasta_dict = genomic.read_fasta(sequence_buf)
a = asp.asp()
a.load_model(modelfname)
(acc_result, don_result) = _predict(a, sequence_buf, (start, stop),
strand)
if feature == 'donor':
return HttpResponse(json.dumps(don_result))
else:
return HttpResponse(json.dumps(acc_result))
except ValueError as e:
return HttpResponse(json.dumps({"status": e.message}))
except:
import traceback
return HttpResponse(
json.dumps({"status": repr(traceback.format_exc())}))
def _asp(request):
try:
organism = str(request.POST['organism'])
sequence = str(request.POST['sequence'])
start = int(request.POST['start'])
stop = int(request.POST['stop'])
strand = str(request.POST['strand'])
feature = str(request.POST['feature'])
if strand == '-':
strand = '-'
else:
strand = '+'
modelfname = 'data/asp/%s.dat.bz2' % organism
# length limit to no smaller than 10 byte and no bigger than 2M
if len(sequence) <= 10 or len(sequence) > 2*1024*1024:
raise ValueError("fasta file incorrect")
if not os.path.isfile(modelfname):
raise ValueError("model should be one of Cress, Fish, Fly, Human, Worm")
if start<80:
raise ValueError("--start value must be >=80")
if stop > 0 and start >= stop - 80:
raise ValueError("--stop value must be > start + 80")
if stop < 0 and stop > -80:
raise ValueError("--stop value must be <= - 80")
# shift the start and stop a bit
start -= 1
stop -= 1
sequence_buf = StringIO.StringIO()
sequence_buf.write(sequence)
sequence_buf.seek(0)
fasta_dict = genomic.read_fasta(sequence_buf)
a = asp.asp()
a.load_model(modelfname)
(acc_result, don_result) = _predict(a, sequence_buf, (start,stop), strand)
if feature == 'donor':
return HttpResponse(json.dumps(don_result))
else:
return HttpResponse(json.dumps(acc_result))
except ValueError as e:
return HttpResponse(json.dumps({"status": e.message}))
except:
import traceback
return HttpResponse(json.dumps({"status": repr(traceback.format_exc())}))
else:
p = [i + 1 for i in preds[1]]
pos = array.array('i')
pos.fromlist(p)
out.tofile(binary_out)
pos.tofile(binary_pos)
def predict_file(self,
fname,
(start, end),
output_format,
score_type,
strand='+'):
skipheader = False
fasta_dict = genomic.read_fasta(file(fname))
if strand == '-':
for k, kseq in fasta_dict.ordered_items():
fasta_dict[k] = genomic.reverse_complement(kseq)
sys.stdout.write('found fasta file with ' + ` len(fasta_dict) ` +
' sequence(s) (strand=%s)\n' % strand)
seqs = seqdict.seqdict(fasta_dict, (start, end))
#get donor/acceptor signal predictions for all sequences
self.signal.predict_acceptor_sites_from_seqdict(seqs)
self.signal.predict_donor_sites_from_seqdict(seqs)
contig_no = 0
for seq in seqs:
return HttpResponse(json.dumps(don_result))
else:
return HttpResponse(json.dumps(acc_result))
except ValueError as e:
return HttpResponse(json.dumps({"status": e.message}))
except:
import traceback
return HttpResponse(
json.dumps({"status": repr(traceback.format_exc())}))
def _predict(asp, sequence_buf, (start, end), strand):
skipheader = False
sequence_buf.seek(0)
fasta_dict = genomic.read_fasta(sequence_buf)
if strand == '-':
for k, kseq in fasta_dict.ordered_items():
fasta_dict[k] = genomic.reverse_complement(kseq)
import seqdict
seqs = seqdict.seqdict(fasta_dict, (start, end))
asp.signal.predict_acceptor_sites_from_seqdict(seqs)
asp.signal.predict_donor_sites_from_seqdict(seqs)
contig_no = 0
for seq in seqs:
contig_no = contig_no + 1
l = len(seq.preds['donor'].get_positions())
(acc_result, don_result) = _predict(a, sequence_buf, (start,stop), strand)
if feature == 'donor':
return HttpResponse(json.dumps(don_result))
else:
return HttpResponse(json.dumps(acc_result))
except ValueError as e:
return HttpResponse(json.dumps({"status": e.message}))
except:
import traceback
return HttpResponse(json.dumps({"status": repr(traceback.format_exc())}))
def _predict(asp, sequence_buf, (start,end), strand):
skipheader = False
sequence_buf.seek(0)
fasta_dict = genomic.read_fasta(sequence_buf)
if strand == '-':
for k, kseq in fasta_dict.ordered_items():
fasta_dict[k]=genomic.reverse_complement(kseq)
import seqdict
seqs = seqdict.seqdict(fasta_dict, (start, end))
asp.signal.predict_acceptor_sites_from_seqdict(seqs)
asp.signal.predict_donor_sites_from_seqdict(seqs)
contig_no = 0;
for seq in seqs:
contig_no = contig_no + 1
l = len(seq.preds['donor'].get_positions())
p=[i-1 for i in preds[1]]
else:
if strand=='+':
p=[i for i in preds[1]]
else:
p=[i+1 for i in preds[1]]
pos=array.array('i')
pos.fromlist(p) ;
out.tofile(binary_out)
pos.tofile(binary_pos)
def predict_file(self, fname, (start,end), output_format, score_type, strand='+'):
skipheader=False
fasta_dict = genomic.read_fasta(file(fname))
if strand=='-':
for k, kseq in fasta_dict.ordered_items():
fasta_dict[k]=genomic.reverse_complement(kseq)
sys.stdout.write('found fasta file with ' + `len(fasta_dict)` + ' sequence(s) (strand=%s)\n' % strand)
seqs= seqdict.seqdict(fasta_dict, (start,end))
#get donor/acceptor signal predictions for all sequences
self.signal.predict_acceptor_sites_from_seqdict(seqs)
self.signal.predict_donor_sites_from_seqdict(seqs)
contig_no = 0 ;
for seq in seqs:
contig_no = contig_no + 1
