def from_file(cls, name, fasta_file, force=False):
'''
Create a Fasta object from a file.
'''
self = cls(name)
with RawFile(fasta_file) as IN, self._db as db:
cur = db.cursor()
cur_chrom = None
seqs = []
name, attrs = None, None
for line in IN:
line = line.strip()
if line.startswith('>'):
# Finish the last chromosome before adding a new one
if len(seqs) > 0:
cur_chrom = Chromosome(name, seqs, *attrs)
self.add_chrom(cur_chrom, cur=cur, force=force)
seqs = []
name, *attrs = line.lstrip('>').split()
else:
seqs += line
#cur_chrom.seq = np.append(cur_chrom.seq,list(line))
# Add the last chromosome
cur_chrom = Chromosome(name, seqs, *attrs)
self.add_chrom(cur_chrom, cur=cur, force=force)
return self
def __getitem__(self, chrom_name):
if chrom_name not in self:
raise ValueError(f'{chrom_name} not in {self._m80_name}')
try:
seq_array = self._bcolz_array(chrom_name)
except Exception as e:
chrom_name = self._get_nickname(chrom_name)
seq_array = self._bcolz_array(chrom_name)
finally:
attrs = [
x[0] for x in self._db.cursor().execute(
'''
SELECT attribute FROM attributes
WHERE chrom = ?
ORDER BY rowid -- This preserves the ordering of attrs
''', (chrom_name, ))
]
return Chromosome(chrom_name, seq_array, *attrs)
def chr1():
return Chromosome('chr1', 'A' * 500000)
def test_init_from_seq():
x = Chromosome('chr1', ['a', 'c', 'g', 't'])
assert True
def chr2():
return Chromosome('chr1', 'C' * 500000)