def test_split_fasta_diff_num_seqs_per_file_alt(self):
"""split_fasta funcs always catches all seqs
"""
# start with 59 seqs (b/c it's prime, so should make more
# confusing splits)
in_seqs = LoadSeqs(data=[('seq%s' % k, 'AACCTTAA') for k in range(59)])
infile = in_seqs.toFasta().split('\n')
# test seqs_per_file from 1 to 1000
for i in range(1, 1000):
_, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(_)
actual = split_fasta(infile, i, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
# remove the files now, so if the test fails they still get
# cleaned up
remove_files(actual)
# building seq collections from infile and the split files result in
# equivalent seq collections
self.assertEqual(LoadSeqs(data=infile, aligned=False),
LoadSeqs(data=actual_seqs, aligned=False))
def test_split_fasta_diff_num_seqs_per_file_alt(self):
"""split_fasta funcs always catches all seqs
"""
# start with 59 seqs (b/c it's prime, so should make more
# confusing splits)
in_seqs = LoadSeqs(data=[('seq%s' % k, 'AACCTTAA') for k in range(59)])
infile = in_seqs.toFasta().split('\n')
# test seqs_per_file from 1 to 1000
for i in range(1, 1000):
_, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(_)
actual = split_fasta(infile, i, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
# remove the files now, so if the test fails they still get
# cleaned up
remove_files(actual)
# building seq collections from infile and the split files result in
# equivalent seq collections
self.assertEqual(
LoadSeqs(data=infile, aligned=False),
LoadSeqs(data=actual_seqs, aligned=False))
def test_split_fasta_diff_num_seqs_per_file(self):
"""split_fasta funcs as expected when diff num seqs go to each file
"""
_, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(_)
infile = [
'>seq1', 'AACCTTAA', '>seq2', 'TTAACC', 'AATTAA', '>seq3',
'CCTT--AA'
]
actual = split_fasta(infile, 2, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
remove_files(actual)
expected = ['%s.%d.fasta' % (filename_prefix, i) for i in range(2)]
# list of file paths is as expected
self.assertEqual(actual, expected)
# building seq collections from infile and the split files result in
# equivalent seq collections
self.assertEqual(LoadSeqs(data=infile, aligned=False),
LoadSeqs(data=actual_seqs, aligned=False))
def test_split_fasta_diff_num_seqs_per_file(self):
"""split_fasta funcs as expected when diff num seqs go to each file
"""
_, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(_)
infile = ['>seq1', 'AACCTTAA', '>seq2', 'TTAACC', 'AATTAA',
'>seq3', 'CCTT--AA']
actual = split_fasta(infile, 2, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
remove_files(actual)
expected = ['%s.%d.fasta' % (filename_prefix, i) for i in range(2)]
# list of file paths is as expected
self.assertEqual(actual, expected)
# building seq collections from infile and the split files result in
# equivalent seq collections
self.assertEqual(
LoadSeqs(data=infile, aligned=False),
LoadSeqs(data=actual_seqs, aligned=False))
def test_split_fasta_equal_num_seqs_per_file(self):
"""split_fasta funcs as expected when equal num seqs go to each file
"""
fd, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(fd)
infile = [
'>seq1', 'AACCTTAA', '>seq2', 'TTAACC', 'AATTAA', '>seq3',
'CCTT--AA'
]
actual = split_fasta(infile, 1, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
remove_files(actual)
expected = ['%s.%d.fasta' % (filename_prefix, i) for i in range(3)]
self.assertEqual(actual, expected)
self.assertEqual(
SequenceCollection.from_fasta_records(parse_fasta(infile), DNA),
SequenceCollection.from_fasta_records(parse_fasta(actual_seqs),
DNA))
def _split_fasta(self, input_fp, params, jobs_to_start, job_prefix, output_dir):
# compute the number of sequences that should be included in
# each file after splitting the input fasta file
num_seqs_per_file = self._compute_seqs_per_file(input_fp, jobs_to_start)
# split the fasta files and get the list of resulting files
tmp_fasta_fps = split_fasta(open(input_fp), num_seqs_per_file, job_prefix, working_dir=output_dir)
return tmp_fasta_fps, True
def _split_fasta(self, input_fp, params, jobs_to_start, job_prefix,
output_dir):
# compute the number of sequences that should be included in
# each file after splitting the input fasta file
num_seqs_per_file = self._compute_seqs_per_file(
input_fp, jobs_to_start)
# split the fasta files and get the list of resulting files
tmp_fasta_fps =\
split_fasta(open(input_fp), num_seqs_per_file,
job_prefix, working_dir=output_dir)
return tmp_fasta_fps, True
def test_split_fasta_equal_num_seqs_per_file(self):
"""split_fasta funcs as expected when equal num seqs go to each file
"""
_, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(_)
infile = ['>seq1', 'AACCTTAA', '>seq2', 'TTAACC', 'AATTAA',
'>seq3', 'CCTT--AA']
actual = split_fasta(infile, 1, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
remove_files(actual)
expected = ['%s.%d.fasta' % (filename_prefix, i) for i in range(3)]
self.assertEqual(actual, expected)
self.assertEqual(
LoadSeqs(data=infile, aligned=False),
LoadSeqs(data=actual_seqs, aligned=False))
def test_split_fasta_equal_num_seqs_per_file(self):
"""split_fasta funcs as expected when equal num seqs go to each file
"""
fd, filename_prefix = mkstemp(dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='')
close(fd)
infile = ['>seq1', 'AACCTTAA', '>seq2', 'TTAACC', 'AATTAA',
'>seq3', 'CCTT--AA']
actual = split_fasta(infile, 1, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
remove_files(actual)
expected = ['%s.%d.fasta' % (filename_prefix, i) for i in range(3)]
self.assertEqual(actual, expected)
self.assertEqual(
SequenceCollection.from_fasta_records(parse_fasta(infile), DNA),
SequenceCollection.from_fasta_records(parse_fasta(actual_seqs), DNA))
def test_split_fasta_equal_num_seqs_per_file(self):
"""split_fasta funcs as expected when equal num seqs go to each file
"""
filename_prefix = get_tmp_filename(tmp_dir=get_qiime_temp_dir(),
prefix='split_fasta_tests',
suffix='',
result_constructor=str)
infile = ['>seq1','AACCTTAA','>seq2','TTAACC','AATTAA',\
'>seq3','CCTT--AA']
actual = split_fasta(infile, 1, filename_prefix)
actual_seqs = []
for fp in actual:
actual_seqs += list(open(fp))
remove_files(actual)
expected = ['%s.%d.fasta' % (filename_prefix,i) for i in range(3)]
self.assertEqual(actual,expected)
self.assertEqual(\
LoadSeqs(data=infile,aligned=False),\
LoadSeqs(data=actual_seqs,aligned=False))
