def test_write_fasta(self):
'''Test write_fasta'''
s = seed.Seed(seq='GAAGGCGGCAGC')
tmpfile = 'tmp.seed.fa'
s.write_fasta(tmpfile, 'spam')
self.assertTrue(filecmp.cmp(tmpfile, os.path.join(data_dir, 'seed_test.write_fasta.fa'), shallow=False))
os.unlink(tmpfile)
def test_extend_with_reads_as_single_end(self):
'''Test _extend_with_reads_as_single_end'''
s = seed.Seed(seq='AGGCT', ext_min_cov=1, verbose=2)
reads1 = os.path.join(data_dir, 'kcount_test.reads_1.fasta')
reads2 = os.path.join(data_dir, 'kcount_test.reads_2.fasta')
s._extend_with_reads_as_single_end(reads1, reads2)
self.assertEqual('TGAGGCTAT', s.seq)
def test_extensions_from_reads_file(self):
'''Test _extensions_from_reads_file'''
s = seed.Seed(seq='AGGCT')
l, r = s._extensions_from_reads_file(os.path.join(data_dir, 'kcount_test.reads_1.fasta'))
self.assertListEqual(l, [])
self.assertListEqual(r, ['A', 'AT', 'AT'])
l, r = s._extensions_from_reads_file(os.path.join(data_dir, 'kcount_test.reads_2.fasta'))
self.assertListEqual(l, ['G', 'TG', 'TG'])
self.assertListEqual(r, [])
def add_new_seed_contig(self, reads1, reads2, contig_name=None, max_attempts=10):
if len(self.contigs):
tmpdir = tempfile.mkdtemp(prefix='tmp.make_seed.', dir=os.getcwd())
tmp_prefix = os.path.join(tmpdir, 'out')
seed_reads1 = tmp_prefix + '_1.fa'
seed_reads2 = tmp_prefix + '_2.fa'
if contig_name is not None:
self._map_reads(reads1, reads2, tmp_prefix, required_flag=5, exclude_flag=8, mate_ref=contig_name)
mapping.bam_to_fasta(tmp_prefix + '.bam', seed_reads1)
seed_reads2 = None
else:
self._get_unmapped_pairs(reads1, reads2, tmp_prefix)
else:
seed_reads1 = reads1
seed_reads2 = reads2
made_seed = False
for i in range(max_attempts):
s = seed.Seed(reads1=seed_reads1, reads2=seed_reads2, extend_length=self.seed_ext_max_bases, seed_length=self.seed_start_length, seed_min_count=self.seed_min_kmer_count, seed_max_count=self.seed_max_kmer_count, ext_min_cov=self.seed_min_cov, ext_min_ratio=self.seed_min_ratio, verbose=self.verbose, kmc_threads=self.kmc_threads, map_threads=self.threads, sequences_to_ignore=self.used_seeds, contigs_to_check=self.contigs)
if s.seq is None or len(s.seq) == 0:
break
if self.seed_overlap_length is None:
s.overlap_length = len(s.seq)
else:
s.overlap_length = self.seed_overlap_length
s.extend(reads1, reads2, self.seed_stop_length)
self.used_seeds.add(s.seq)
if len(s.seq) >= 0.75 * self.seed_stop_length:
made_seed = True
break
elif self.verbose:
print(" Couldn't extend seed enough. That was attempt", i+1, 'of', max_attempts, flush=True)
if len(self.contigs):
shutil.rmtree(tmpdir)
if not made_seed or len(s.seq) == 0:
return None
if self.verbose:
print(" Extended seed OK.", flush=True)
new_name = 'seeded.' + '1'.zfill(5)
i = 1
while new_name in self.contigs:
i += 1
new_name = 'seeded.' + str(i).zfill(5)
self._add_contig(pyfastaq.sequences.Fasta(new_name, s.seq))
return new_name
def test_extension_from_read(self):
'''Test _test_extension_from_read'''
s = seed.Seed(seq='AGGCT')
self.assertEqual(None, s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AAAAA')))
self.assertEqual(None, s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AGGC')))
self.assertEqual('A', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AGGCTA')))
self.assertEqual('AT', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AGGCTAT')))
self.assertEqual('AT', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'GGGAGGCTAT')))
self.assertEqual('AA', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'TTAGCCT')))
self.assertEqual(None, s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AAAAA'), left=True))
self.assertEqual(None, s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AGGCTA'), left=True))
self.assertEqual('GT', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'GTAGGCTA'), left=True))
self.assertEqual('GT', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'GTAGGCTATTC'), left=True))
self.assertEqual('GT', s._extension_from_read(pyfastaq.sequences.Fasta('x', 'AGCCTAC'), left=True))
def _make_new_seed(self, seed_name):
if self.verbose:
print('Making new seed for', seed_name, ' ... start')
tmp_prefix = os.path.join(self.tmpdir, 'out')
seed_reads = tmp_prefix + '.' + seed_name + '.reads_1.fa'
if len(self.original_seeds[seed_name]) > self.seed_stop_length:
start = int(0.5 * len(self.original_seeds[seed_name]) -
0.5 * self.seed_stop_length)
end = int(0.5 * len(self.original_seeds[seed_name]) +
0.5 * self.seed_stop_length)
else:
start = None
end = None
if self.verbose:
print('Making new seed for', seed_name, ' ... getting reads')
mapping.bam_file_to_region_fasta(self.bam_file, seed_reads, seed_name,
start, end)
if self.verbose:
print('Making new seed for', seed_name,
' ... finding most common kmer')
new_seed = seed.Seed(extend_length=self.extend_length,
overlap_length=self.overlap_length,
reads1=seed_reads,
ext_min_cov=self.ext_min_cov,
ext_min_ratio=self.ext_min_ratio,
verbose=self.verbose,
seed_length=self.seed_length,
seed_min_count=self.seed_min_count,
seed_max_count=self.seed_max_count,
kmc_threads=self.kmc_threads,
map_threads=self.threads)
if len(new_seed) == 0:
print('Warning: could not get most common kmer for', seed_name)
return
if self.verbose:
print('Making new seed for', seed_name,
' ... extending most common kmer')
new_seed.extend(self.reads1, self.reads2, self.seed_stop_length)
f = pyfastaq.utils.open_file_write(tmp_prefix + '.' + seed_name +
'.fa')
print(pyfastaq.sequences.Fasta('seed.' + seed_name,
new_seed.seq[10:-10]),
file=f)
pyfastaq.utils.close(f)
if self.verbose:
print('Making new seed for', seed_name, ' ... finished')
def test_len(self):
'''Test len'''
s = seed.Seed(seq='AGGCT')
self.assertEqual(5, len(s))
s.seq = None
self.assertEqual(0, len(s))