def from_gff3(path, attributes=None, region=None, score_fill=-1,
phase_fill=-1, attributes_fill=b'.', dtype=None,
blen=None, storage=None, create='table', expectedlen=200000,
**storage_kwargs):
# setup iterator
recs = iter_gff3(path, attributes=attributes, region=region,
score_fill=score_fill, phase_fill=phase_fill,
attributes_fill=attributes_fill)
# read a sample to determine dtype, blen
recs_sample = list(itertools.islice(recs, 1000))
names = 'seqid', 'source', 'type', 'start', 'end', 'score', 'strand', \
'phase'
if attributes:
names += tuple(attributes)
ra = np.rec.array(recs_sample, names=names, dtype=dtype)
dtype = ra.dtype
# setup output
storage = _chunked.get_storage(storage)
out = getattr(storage, create)(ra, expectedlen=expectedlen,
**storage_kwargs)
blen = _chunked.get_blen_table(out, blen=blen)
# read block-wise
block = list(itertools.islice(recs, 0, blen))
while block:
a = np.asarray(block, dtype=dtype)
out.append(a)
block = list(itertools.islice(recs, 0, blen))
out = FeatureChunkedTable(out)
return out
def to_vcf(self,
path,
rename=None,
number=None,
description=None,
fill=None,
blen=None,
write_header=True):
names, callset = normalize_callset(self)
with open(path, 'w') as vcf_file:
if write_header:
write_vcf_header(vcf_file,
names,
callset,
rename=rename,
number=number,
description=description)
blen = _chunked.get_blen_table(self, blen)
for i in range(0, len(self), blen):
j = min(i + blen, len(self))
block = self[i:j]
write_vcf_data(vcf_file,
names,
block,
rename=rename,
fill=fill)
def to_vcf(self, path, rename=None, number=None, description=None,
fill=None, blen=None, write_header=True):
with open(path, 'w') as vcf_file:
if write_header:
write_vcf_header(vcf_file, self, rename=rename, number=number,
description=description)
blen = _chunked.get_blen_table(self)
for i in range(0, len(self), blen):
j = min(i+blen, len(self))
block = self[i:j]
write_vcf_data(vcf_file, block, rename=rename, fill=fill)
def from_gff3(path,
attributes=None,
region=None,
score_fill=-1,
phase_fill=-1,
attributes_fill=b'.',
dtype=None,
blen=None,
storage=None,
create='table',
expectedlen=200000,
**storage_kwargs):
# setup iterator
recs = iter_gff3(path,
attributes=attributes,
region=region,
score_fill=score_fill,
phase_fill=phase_fill,
attributes_fill=attributes_fill)
# read a sample to determine dtype, blen
recs_sample = list(itertools.islice(recs, 1000))
names = 'seqid', 'source', 'type', 'start', 'end', 'score', 'strand', \
'phase'
if attributes:
names += tuple(attributes)
ra = np.rec.array(recs_sample, names=names, dtype=dtype)
dtype = ra.dtype
# setup output
storage = _chunked.get_storage(storage)
out = getattr(storage, create)(ra,
expectedlen=expectedlen,
**storage_kwargs)
blen = _chunked.get_blen_table(out, blen=blen)
# read block-wise
block = list(itertools.islice(recs, 0, blen))
while block:
a = np.asarray(block, dtype=dtype)
out.append(a)
block = list(itertools.islice(recs, 0, blen))
out = FeatureChunkedTable(out)
return out
