def check_kmer_overlap(f):
chr2 = f['chr2']
kmers = Fasta.as_kmers(chr2, 10, overlap=2)
for i, k in enumerate(list(kmers)[:-1]):
assert (len(k[1]) == 10)
assert (k[0] == (i * (10 - 2)))
kmers = Fasta.as_kmers(chr2, 10, overlap=4)
seqs = [k[1] for k in kmers]
paired_seqs = zip(seqs[0:-1], seqs[1:])
for a, b in paired_seqs:
if len(a) < 4 or len(b) < 4: continue
assert (a[-4:] == b[:4])
def segments(self):
'''
Generator for Segments
'''
startchr = self.start_chromosome
start = self.start_location
chrs = [
x[0]
for x in sorted(self.fasta.index.items(), key=lambda a: a[1][0])
]
for chr in chrs:
segcount = 0
if self.verbose:
print "Reading chr %s" % chr
# Skip forward if a starting chr was defined
if startchr is not None and startchr != chr:
continue
else:
startchr = None
for kmer in Fasta.as_kmers(self.fasta[chr], self.segment_size):
end = start + self.segment_size
seg = Segment(start, end, kmer[1], chr)
segcount += 1
if self.verbose and segcount % 1000 == 0:
print "Read %d segments" % segcount
yield seg
start = end
def segments(self):
'''
Generator for Segments
'''
startchr = self.start_chromosome
start = self.start_location
chrs = [x[0] for x in sorted(self.fasta.index.items(), key=lambda a: a[1][0])]
for chr in chrs:
segcount = 0
if self.verbose:
print "Reading chr %s" % chr
# Skip forward if a starting chr was defined
if startchr is not None and startchr != chr:
continue
else:
startchr = None
for kmer in Fasta.as_kmers(self.fasta[chr],self.segment_size):
end = start + self.segment_size
seg = Segment(start, end, kmer[1] ,chr)
segcount += 1
if self.verbose and segcount % 1000 == 0:
print "Read %d segments" % segcount
yield seg
start = end
def check_kmers(f):
seq = str(f['chr2'])
kmers = list(Fasta.as_kmers(f['chr2'], 10))
assert (len(kmers) == len(seq) / 10)
assert (kmers[0] == (0, seq[:10]))
seqs = [k[1] for k in kmers]
assert ("".join(seqs) == seq)
last_pair = kmers[-1]
assert (seqs[-1][-1] == 'T')
seq = str(f['chr3'])
kmers = list(Fasta.as_kmers(f['chr3'], 1))
assert (kmers[2][0] == 2)
seqs = [k[1] for k in kmers]
assert ("".join(seqs) == seq)
def with_kmers(f, names, k, overlap):
"""
split the sequences in Fasta object `f` into pieces of length `k`
with the given `overlap` the results are written to the array of files
`fhs`
"""
fhs = [open(name, 'wb') for name in names]
i = 0
for seqid in f.keys():
seq = f[seqid]
for (start0, subseq) in Fasta.as_kmers(seq, k, overlap=overlap):
fh = fhs[i % len(fhs)]
print >>fh, ">%s" % format_kmer(seqid, start0)
print >>fh, subseq
i += 1
def with_kmers(f, names, k, overlap):
"""
split the sequences in Fasta object `f` into pieces of length `k`
with the given `overlap` the results are written to the array of files
`fhs`
"""
fhs = [open(name, 'wb') for name in names]
i = 0
for seqid in f.keys():
seq = f[seqid]
for (start0, subseq) in Fasta.as_kmers(seq, k, overlap=overlap):
fh = fhs[i % len(fhs)]
print >> fh, ">%s" % format_kmer(seqid, start0)
print >> fh, subseq
i += 1
