def __init__(self,
target_files_dict,
num_threads_to_use=1,
progress=progress,
run=run):
"""A class to streamline HMM runs."""
self.num_threads_to_use = num_threads_to_use
self.progress = progress
self.run = run
self.tmp_dirs = []
self.target_files_dict = {}
for source in target_files_dict:
tmp_dir = filesnpaths.get_temp_directory_path()
self.tmp_dirs.append(tmp_dir)
part_file_name = os.path.join(
tmp_dir, os.path.basename(target_files_dict[source]))
# create splitted fasta files inside tmp directory
self.target_files_dict[source] = utils.split_fasta(
target_files_dict[source],
parts=self.num_threads_to_use,
prefix=part_file_name)
def __init__(self, target_files_dict, num_threads_to_use=1, program_to_use='hmmscan', progress=progress, run=run):
"""A class to streamline HMM runs.
Notes
=====
- HMMer user guide: http://eddylab.org/software/hmmer/Userguide.pdf
"""
self.num_threads_to_use = num_threads_to_use
self.program_to_use = program_to_use
self.progress = progress
self.run = run
self.tmp_dirs = []
self.target_files_dict = {}
acceptable_programs = ["hmmscan", "hmmsearch"]
if self.program_to_use not in acceptable_programs:
raise ConfigError("HMMer class here. You are attempting to use the program %s to run HMMs, but we don't recognize it. The currently "
"supported programs are: %s" % (self.program_to_use, ", ".join(acceptable_programs)))
for source in target_files_dict:
tmp_dir = filesnpaths.get_temp_directory_path()
self.tmp_dirs.append(tmp_dir)
part_file_name = os.path.join(tmp_dir, os.path.basename(target_files_dict[source]))
# create splitted fasta files inside tmp directory
self.target_files_dict[source] = utils.split_fasta(target_files_dict[source],
parts=self.num_threads_to_use,
prefix=part_file_name)
def test_custom_prefix(self):
parts = 1
prefix = os.path.join(self.this_dir, 'silly')
out_files = split_fasta(self.five_seq_fasta,
parts=parts,
prefix=prefix)
expected_out_files = [os.path.join(self.this_dir, 'silly.0')]
self.assertEqual(out_files, expected_out_files)
for f in out_files:
os.remove(f)
def test_more_parts_than_sequences(self):
parts = 10
num_sequences = 5
expected_out_files = [
os.path.join(self.test_files, f'{self.five_seq_fasta}.{i}')
for i in range(num_sequences)
]
out_files = split_fasta(self.five_seq_fasta, parts=parts)
self.assertEqual(out_files, expected_out_files)
for f in out_files:
os.remove(f)
def test_custom_prefix(self):
parts = 1
file_name_prefix = 'silly'
out_files = split_fasta(self.five_seq_fasta,
parts=parts,
file_name_prefix=file_name_prefix,
output_dir=self.this_dir)
expected_out_files = [os.path.join(self.this_dir, 'silly.0')]
self.assertEqual(out_files, expected_out_files)
for f in out_files:
os.remove(f)
def test_shuffle_mode(self):
parts = 2
out_files = split_fasta(self.five_seq_fasta, parts=parts, shuffle=True)
fasta = ReadFasta(out_files[0])
self.assertEqual(fasta.ids, ['seq1 apple', 'seq3 cat', 'seq5 extra'])
self.assertEqual(fasta.sequences, ['AA', 'ACTACT', 'ACTGAACTGA'])
fasta.close()
fasta = ReadFasta(out_files[1])
self.assertEqual(fasta.ids, ['seq2 banana', 'seq4 dog'])
self.assertEqual(fasta.sequences, ['ACAC', 'ACTGACTG'])
fasta.close()
for f in out_files:
os.remove(f)
def test_single_fasta_gives_one_split(self):
out_files = split_fasta(self.single_seq_fasta)
expected_out_file = os.path.join(self.test_files,
f'{self.single_seq_fasta}.0')
self.assertEqual(out_files, [expected_out_file])
self.assertTrue(os.path.exists(expected_out_file))
fasta = ReadFasta(expected_out_file)
self.assertEqual(fasta.ids, ['seq1 apple'])
self.assertEqual(fasta.sequences, ['AA'])
fasta.close()
os.remove(expected_out_file)
def _split_input_file(self):
"""Split input fasta into the correct number of splits.
Returns `State` with the paths to each of the fasta splits.
Raises `ValueError` if `self.number_of_splits < 0.
See Superclass `_split_input_file` for more info.
"""
# Todo: probably should move this check into the constructor.
if self.number_of_splits <= 0:
ValueError(
f'number_of_splits muts be > 0. Got {self.number_of_splits}')
# Todo are there errors to catch here?
self.input_file_splits = utils.split_fasta(self.input_file_path,
parts=self.number_of_splits,
shuffle=True)
return State(input_file_splits=self.input_file_splits)
def test_fasta_splitting(self):
parts = 2
expected_out_files = [
os.path.join(self.test_files, f'{self.five_seq_fasta}.{i}')
for i in range(parts)
]
out_files = split_fasta(self.five_seq_fasta, parts=parts)
self.assertEqual(out_files, expected_out_files)
fasta = ReadFasta(out_files[0])
self.assertEqual(fasta.ids, ['seq1 apple', 'seq2 banana'])
self.assertEqual(fasta.sequences, ['AA', 'ACAC'])
fasta.close()
fasta = ReadFasta(out_files[1])
self.assertEqual(fasta.ids, ['seq3 cat', 'seq4 dog', 'seq5 extra'])
self.assertEqual(fasta.sequences, ['ACTACT', 'ACTGACTG', 'ACTGAACTGA'])
fasta.close()
for f in out_files:
os.remove(f)