def _region_to_seqs(self, track, extend_up=0, extend_down=0):
BUFSIZE = 10000
if isinstance(track, list):
for name in track:
chrom, coords = name.split(":")
start, end = [int(c) for c in coords.split("-")]
start += 1
start -= extend_up
end += extend_down
seq = self.get_seq(chrom, start, end)
yield Sequence(name, seq.seq)
else:
with open(track) as fin:
lines = fin.readlines(BUFSIZE)
while lines:
for line in lines:
name = line.strip()
chrom, coords = name.split(":")
start, end = [int(c) for c in coords.split("-")]
start += 1
start -= extend_up
end += extend_down
seq = self.get_seq(chrom, start, end)
yield Sequence(name, seq.seq)
lines = fin.readlines(BUFSIZE)
def _variant_to_sequence(variants):
"""
Convert `cyvcf2.Variant` objects to `pyfaidx.Seqeunce` objects
for reference and variants.
"""
for v in variants:
ref = Sequence(name=v.chrom, seq=v.ref,
start=v.start, end=v.start + len(v.ref))
alt = Sequence(name=v.chrom, seq=v.alt,
start=v.start, end=v.start + len(v.alt))
yield ref, alt
def test_interval_seq_builder_concat(interval_seq_builder):
with pytest.raises(TypeError):
interval_seq_builder.concat()
sequence = Sequence(seq='CCCCATCGNN', start=10, end=20)
interval_seq_builder.restore(sequence)
assert interval_seq_builder.concat() == 'CCCCTAGCNN'
def interval_seq_builder():
return IntervalSeqBuilder([
Interval('chr1', 10, 13),
Interval('chr1', 13, 14),
Sequence(seq='TAGC', start=14, end=18),
Interval('chr1', 18, 20)
])
def _bed_to_seqs(self, track, stranded=False, extend_up=0, extend_down=0):
BUFSIZE = 10000
with open(track) as fin:
lines = fin.readlines(BUFSIZE)
while lines:
for line in lines:
if line.startswith("#") or line.startswith("track"):
continue
vals = line.strip().split("\t")
try:
start, end = int(vals[1]), int(vals[2])
except ValueError:
raise
rc = False
if stranded:
try:
rc = vals[5] == "-"
except IndexError:
pass
starts = [start]
ends = [end]
chrom = vals[0]
# BED12
if len(vals) == 12:
starts = [int(x) for x in vals[11].split(",")[:-1]]
sizes = [int(x) for x in vals[10].split(",")[:-1]]
starts = [start + x for x in starts]
ends = [
start + size for start, size in zip(starts, sizes)
]
name = "{}:{}-{}".format(chrom, start, end)
try:
name = " ".join((name, vals[3]))
except Exception:
pass
starts = [start + 1 for start in starts]
# extend
if extend_up:
if rc:
ends[-1] += extend_up
else:
starts[0] -= extend_up
if extend_down:
if rc:
starts[0] -= extend_down
else:
ends[-1] += extend_down
intervals = zip(starts, ends)
seq = self.get_spliced_seq(chrom, intervals, rc)
yield Sequence(name, seq.seq)
lines = fin.readlines(BUFSIZE)
def locate(args):
kmers, fd, fo = args.kmer, args.db, args.out
fg = args.fg
db = Fasta(fd)
#
kseqs = kmers.split(',')
kseqs2 = [Sequence(name='kmer',seq=kseq).reverse.complement.seq for kseq in kseqs]
ptn = "|".join([ "("+k+")" for k in kseqs+kseqs2 ])
#
seqs = []
if fg != '':
fhg = open(fg, 'r')
for line in fhg:
line = line.rstrip("\n")
if not line: continue
gid = line.split()[0]
if gid == 'gid': continue
if gid not in db: continue
seqs.append(gid)
else:
seqs = db.keys()
fho = open(fo, 'w')
fho.write('kmer\tsid\tstart\tend\tsrd\n')
i = 1
for seqid in seqs:
seq = db[seqid][0:].seq
for m in re.finditer(ptn, seq):
start, end = m.start()+1, m.end()
srd = "+" if m.group(0) in kseqs else "-"
fho.write(f"{m.group(0)}\t{seqid}\t{start}\t{end}\t{srd}\n")
i += 1
fho.close()
def _variant_to_sequence(self, variants):
"""
Convert `cyvcf2.Variant` objects to `pyfaidx.Seqeunce` objects
for reference and variants.
"""
for v in variants:
ref = Sequence(name=v.CHROM,
seq=v.REF,
start=v.start,
end=v.start + len(v.REF))
# TO DO: consider alternative alleles.
alt = Sequence(name=v.CHROM,
seq=v.ALT[0],
start=v.start,
end=v.start + len(v.ALT[0]))
yield ref, alt
def _regions_to_seqs(self, track, extend_up=0, extend_down=0):
if isinstance(track, list):
for region in track:
name = region.strip()
seq = self._region_to_seq(name, extend_up, extend_down)
yield Sequence(name, seq)
else:
with open(track) as fin:
bufsize = 10000
lines = fin.readlines(bufsize)
for region in lines:
name = region.strip()
seq = self._region_to_seq(name, extend_up, extend_down)
yield Sequence(name, seq)
# load more lines if needed
lines += fin.readlines()
def test__split_overlapping(variant_seq_extractor):
pair = (Sequence(seq='AAA', start=3,
end=6), Sequence(seq='T', start=3, end=4))
splited_pairs = list(variant_seq_extractor._split_overlapping([pair], 5))
assert splited_pairs[0][0].seq == 'AA'
assert splited_pairs[0][1].seq == 'T'
assert splited_pairs[1][0].seq == 'A'
assert splited_pairs[1][1].seq == ''
pair = (Sequence(seq='TT', start=3,
end=5), Sequence(seq='AAA', start=3, end=6))
splited_pairs = list(variant_seq_extractor._split_overlapping([pair], 4))
assert splited_pairs[0][0].seq == 'T'
assert splited_pairs[0][1].seq == 'A'
assert splited_pairs[1][0].seq == 'T'
assert splited_pairs[1][1].seq == 'AA'
def _bed_to_seqs(self, track, stranded=False, extend_up=0, extend_down=0):
bufsize = 10000
with open(track) as fin:
lines = fin.readlines(bufsize)
for line in lines:
if line.startswith("#") or line.startswith("track"):
continue
vals = line.strip().split("\t")
chrom, start, end = str(vals[0]), int(vals[1]), int(vals[2])
name = f"{chrom}:{start}-{end}"
# there might be more...
starts = [start]
ends = [end]
# BED4: add name column to name
if len(vals) >= 4:
name = " ".join((name, vals[3]))
# BED5: check strandedness
rc = False
if stranded and len(vals) >= 6:
rc = vals[5] == "-"
# BED12: get all blocks
if len(vals) >= 12:
starts = [int(x) for x in vals[11].split(",")[:-1]]
sizes = [int(x) for x in vals[10].split(",")[:-1]]
starts = [start + x for x in starts]
ends = [start + size for start, size in zip(starts, sizes)]
# convert to 1-based counting
starts = [start + 1 for start in starts]
# extend
if extend_up:
if rc:
ends[-1] += extend_up
else:
starts[0] -= extend_up
if extend_down:
if rc:
starts[0] -= extend_down
else:
ends[-1] += extend_down
intervals = zip(starts, ends)
seq = self.get_spliced_seq(chrom, intervals, rc)
yield Sequence(name, seq.seq)
# load more lines if needed
lines += fin.readlines(1)
def get_spliced_seq(self, name, intervals, rc=False):
"""Return a sequence by record name and list of intervals
Interval list is an iterable of [start, end].
Coordinates are 0-based, end-exclusive.
"""
# Get sequence for all intervals
chunks = [self.faidx.fetch(name, s, e) for s, e in intervals]
start = chunks[0].start
end = chunks[-1].end
# reverce complement
if rc:
seq = "".join([(-chunk).seq for chunk in chunks[::-1]])
else:
seq = "".join([chunk.seq for chunk in chunks])
return Sequence(name=name, seq=seq, start=start, end=end)
def regions_to_seqs(self, track, extend_up=0, extend_down=0):
# if track is a file, loop over its lines
lines = track
if isinstance(track, str):
lines = parse_file(track)
for line in lines:
name = line.split()[0]
try:
if bad_coords(name, track):
continue
seq = region_to_seq(self, name, extend_up, extend_down)
except (ValueError, IndexError):
msg = f"Skipping region that cannot be parsed: '{name}'"
if isinstance(track, str):
msg = f"Skipping region from '{os.path.basename(track)}' that cannot be parsed: '{name}'"
logger.warning(msg)
continue
yield Sequence(name, seq)
def bed_to_seq(self, vals, stranded=False, extend_up=0, extend_down=0):
chrom, start, end = str(vals[0]), int(vals[1]), int(vals[2])
name = f"{chrom}:{start}-{end}"
# there might be more...
starts = [start]
ends = [end]
# BED4: add name column to name
if len(vals) >= 4:
name = " ".join((name, vals[3]))
# BED5: check strandedness
rc = False
if stranded and len(vals) >= 6:
rc = vals[5] == "-"
# BED12: get all blocks
if len(vals) >= 12:
starts = [int(x) for x in vals[11].split(",")[:-1]]
sizes = [int(x) for x in vals[10].split(",")[:-1]]
starts = [start + x for x in starts]
ends = [start + size for start, size in zip(starts, sizes)]
# convert to 1-based counting
starts = [start + 1 for start in starts]
# extend
if extend_up:
if rc:
ends[-1] += extend_up
else:
starts[0] -= extend_up
if extend_down:
if rc:
starts[0] -= extend_down
else:
ends[-1] += extend_down
intervals = zip(starts, ends)
seq = self.get_spliced_seq(chrom, intervals, rc).seq
return Sequence(name, seq)
def test_interval_seq_builder_restore(interval_seq_builder):
sequence = Sequence(seq='CCCCATCGTT', start=10, end=20)
interval_seq_builder.restore(sequence)
assert interval_seq_builder[0].seq == 'CCC'
assert interval_seq_builder[1].seq == 'C'
assert interval_seq_builder[2].seq == 'TAGC'
assert interval_seq_builder[3].seq == 'TT'
interval_seq_builder.append(Interval('chr1', 5, 10))
interval_seq_builder.restore(sequence)
assert interval_seq_builder[4].seq == ''
interval_seq_builder.append(Interval('chr1', 20, 25))
interval_seq_builder.restore(sequence)
assert interval_seq_builder[5].seq == ''
interval_seq_builder.append(Interval('chr1', 10, 5))
interval_seq_builder.restore(sequence)
assert interval_seq_builder[6].seq == ''
interval_seq_builder.append(Interval('chr1', 25, 20))
interval_seq_builder.restore(sequence)
assert interval_seq_builder[7].seq == ''
def read_kmer(fk, nfea):
kms = []
fhk = open(fk,'r')
for line in fhk:
line = line.rstrip("\n")
if not line: continue
i, opt, bin, epi, pval, fid, fname, kmers = line.split()[:8]
if i == 'i': continue
i = int(i)
if nfea == 'top5' and i > 5: break
if nfea == 'top10' and i > 10: break
if nfea == 'top30' and i > 30: break
if nfea == 'top50' and i > 50: break
if nfea == 'top100' and i > 100: break
if nfea == 'top200' and i > 200: break
if nfea == 'top300' and i > 300: break
if nfea == 'top500' and i > 500: break
kseqs = kmers.split(',')
kseqs2 = [Sequence(name='kmer',seq=kseq).reverse.complement.seq for kseq in kseqs]
ptn = "|".join([ "("+k+")" for k in kseqs+kseqs2 ])
kms.append([fid,ptn,set(kseqs)])
fhk.close()
return kms
from pyfaidx import Sequence, complement
from nose.tools import assert_raises, raises
seq = Sequence(name='gi|557361099|gb|KF435150.1|', seq='TTGAAGATTTTGCATGCAGCAGGTGCGCAAGGTGAAATGTTCACTGTTAAA',
start=100, end=150)
seq_invalid = Sequence(name='gi|557361099|gb|KF435150.1|', seq='TTGAAGATTTPGCATGCAGCAGGTGCGCAAGGTGAAATNTTCACTGTTAAA',
start=100, end=150)
comp_valid = 'TTGAAGATTTnGCATGCAGCAGGtgccaAGGTGAAATGTTNACTGTTAAA'
comp_invalid = 'TTGAAGATTTnGCATGCAGCPQGtgccaAGGTGAAATGTTNACTGTTAAA'
def test_negate():
assert str(-seq) == str(seq.complement[::-1])
def test_negate_metadata():
# Negate should affect __repr__ the same way as reverse and complement
seq_neg = -seq
assert seq_neg.__repr__() == seq.complement[::-1].__repr__()
def test_seq_invalid():
assert_raises(ValueError, lambda: seq_invalid.complement)
def test_integer_index():
assert seq[1].seq == 'T'
def test_slice_index():
assert seq[0:10].seq == 'TTGAAGATTT'
@raises(ValueError)
def test_check_coordinates():
x = Sequence(name='gi|557361099|gb|KF435150.1|', seq='TTGAAGATTTTGCATGCAGCAGGTGCGCAAGGTGAAATGTTCACTGTTAAA',
start=100, end=110)
x[:]
def _fetch(self, interval, istart, iend):
# fetch interval, ignore strand
seq = self.ref_seq_extractor.extract(
Interval(interval.chrom, istart, iend))
seq = Sequence(name=interval.chrom, seq=seq, start=istart, end=iend)
return seq
def _fetch(self, interval, istart, iend):
seq = self.fasta.extract(Interval(interval.chrom, istart, iend))
seq = Sequence(name=interval.chrom, seq=seq, start=istart, end=iend)
return seq
