def load_datasets(self,
path_pos,
path_neg,
background_alg,
SHOULD_SHORT=0,
HEADERS=False,
HARDMASK=False
):
positive = []
negative = []
header_pos = []
header_neg = []
print "HARDMASK:", HARDMASK
#Load positive for path_pos FASTA
for line in self.readfasta(path_pos, HARDMASK):
if HEADERS and line[0] == '>':
header_pos.append(line)
else:
positive.append(line)
if SHOULD_SHORT > 0 and int(SHOULD_SHORT) == len(positive):
break
#Load background as shuffled copy of pisitive
if path_neg != None:
#Load background from path_neg FASTA
for line in self.readfasta(path_neg): # Should bg determi hardmasking with the same variable as positive?
if HEADERS and line[0] == '>':
header_neg.append(line)
else:
negative.append(line)
if SHOULD_SHORT > 0 and int(SHOULD_SHORT) == len(negative):
break
elif background_alg == "dinuclShuffle":
for line in positive:
sh_line = altschulEriksonDinuclShuffle.dinuclShuffle(line.upper())
negative.append(sh_line)
else:
#No background
negative.append("NNNNNNNNNNNNNNNNNNNNNNNNNNN")
print "Lengths:", len(positive), len(negative)
if HEADERS:
return positive, negative, header_pos, header_neg
else:
return positive, negative
def generate_sequences(seqs, nfold):
bg_gc_list = []
bg_lengths = []
for record in seqs:
seq = record.seq.__str__()
for n in range(0, nfold):
new_sequence = ""
for sequence in split_seq(seq):
if re.match("N", sequence):
new_sequence += sequence
elif sequence:
new_sequence += dinuclShuffle(sequence)
new_seq = SeqRecord(Seq(new_sequence, generic_dna),
id="background_seq_for_{0:s}".format(
record.name),
description="")
print(new_seq.format("fasta"), end="")
bg_gc_list.append(GC(new_sequence))
bg_lengths.append(len(new_sequence))
return bg_gc_list, bg_lengths
def generate_sequences(seqs, nfold):
cpt = 1
bg_gc_list = []
bg_lengths = []
for record in seqs:
seq = record.seq.__str__()
descr = "Background sequence for {0:s}".format(record.name)
for n in range(0, nfold):
new_sequence = ""
for sequence in split_seq(seq):
if re.match('N', sequence):
new_sequence += sequence
elif sequence:
new_sequence += dinuclShuffle(sequence)
new_seq = SeqRecord(Seq(new_sequence, generic_dna),
id="background_seq_{0:d}".format(cpt), description=descr)
print new_seq.format("fasta"),
bg_gc_list.append(GC(new_sequence))
bg_lengths.append(len(new_sequence))
cpt += 1
return bg_gc_list, bg_lengths
def main(fileName, NUM):
seq = file2string(fileName)
for i in range(NUM):
shuffledSeq = dinuclShuffle(seq)
sys.stdout.write(">%d\n" % (i + 1))
sys.stdout.write("%s\n" % shuffledSeq)
def main(fileName,NUM):
seq = file2string(fileName)
for i in range(NUM):
shuffledSeq = dinuclShuffle(seq)
sys.stdout.write(">%d\n" % (i+1))
sys.stdout.write("%s\n" % shuffledSeq)
def shuffle_window(ss, wl, step):
bs = ss[:]
for i in range(0, len(bs) - 1, step):
#print i,"\t",ss[i:(i+wl)]
bs = bs[0:i] + dinuclShuffle(bs[i:(i + wl)]) + bs[i + wl:]
return (bs) # returns shuffled sequence
def shuffle_window(ss, wl, step):
bs = ss[:]
for i in range (0, len(bs)-1, step):
#print i,"\t",ss[i:(i+wl)]
bs = bs[0:i] + dinuclShuffle(bs[i:(i+wl)]) + bs[i+wl:]
return(bs) # returns shuffled sequence
