def run(self):
# todo: filter fasta to keep just the largest contig.
# run the fasta through barrnap
fd, barrnap_outputfile = mkstemp()
self.files_to_cleanup.append(barrnap_outputfile)
b = Barrnap(self.input_file, self.threads)
subprocess.check_output(
b.construct_barrnap_command(barrnap_outputfile), shell=True)
boundries = b.read_barrnap_output(barrnap_outputfile)
f = Fasta(self.input_file)
fragments = f.calc_fragment_coords(boundries)
f.populate_fragments_from_chromosome(fragments,
self.max_bases_from_ends)
ff = FragmentFiles(fragments,
self.output_directory,
fragment_order=self.fragment_order)
ff.create_fragment_fastas()
# create a default profile.txt file
default_profile = ProfileGenerator(self.output_directory,
len(ff.ordered_fragments),
self.dnaa_fasta, self.threads)
default_profile.write_output_file()
def test_calc_fragment_coords_gz(self):
f = Fasta(os.path.join(data_dir, 'calc_fragment_coords.fa.gz'), False)
boundries = [[45, 55], [90, 110], [150, 180]]
fragments = f.calc_fragment_coords(boundries)
coords = [f.coords for f in fragments]
self.assertEqual(coords,
[[[180, 200], [0, 45]], [[55, 90]], [[110, 150]]])
def test_chop_from_ends(self):
f = Fasta(os.path.join(data_dir, 'calc_fragment_coords.fa'), False)
fragments = f.calc_fragment_coords([[45, 55], [90, 110], [150, 180]])
sequences = [str(f.sequence) for f in fragments]
f.populate_fragments_from_chromosome(fragments, 5)
sequences = [str(f.sequence) for f in fragments]
self.assertEqual(sequences,
['TTTTTNNNAAAAA', 'CCCCCNNNCCCCC', 'GGGGGNNNGGGGG'])
def test_calc_fragment_coords(self):
f = Fasta(os.path.join(data_dir, 'calc_fragment_coords.fa'), False)
boundries = [[45, 55], [90, 110], [150, 180]]
fragments = f.calc_fragment_coords(boundries)
coords = [f.coords for f in fragments]
self.assertEqual(coords,
[[[180, 200], [0, 45]], [[55, 90]], [[110, 150]]])
f.populate_fragments_from_chromosome(fragments, None)
def run_analysis(self, input_file, p, d):
# run the fasta through barrnap
fd, barrnap_outputfile = mkstemp()
b = Barrnap(input_file, self.threads)
subprocess.check_output(
b.construct_barrnap_command(barrnap_outputfile),
shell=True)
boundries = b.read_barrnap_output(barrnap_outputfile)
f = Fasta(input_file, is_circular = self.is_circular)
fragments = f.calc_fragment_coords( boundries)
f.populate_fragments_from_chromosome(fragments, self.max_bases_from_ends)
tmpdir = mkdtemp()
self.dirs_to_cleanup.append(tmpdir)
ff = FragmentFiles(fragments, tmpdir)
ff.create_fragment_fastas()
# take each fasta file and blast it against the database
blast = Blast(d.db_prefix, self.threads)
gat_profile = GATProfile(fragments = [])
for fasta_file in ff.output_filenames:
blast_results = blast.run_blast(fasta_file)
fb = FilterBlast(blast_results, self.min_bit_score, self.min_alignment_length)
top_result = fb.return_top_result()
if top_result is None:
gat_profile.fragments.append('?')
fasta_file
with open(fasta_file, "r") as fasta_file_fh:
with open(self.new_fragments, "a+") as newfrag_fh:
newfrag_fh.write(fasta_file_fh.read())
continue
else:
self.top_results.append(top_result)
if top_result.is_forward():
gat_profile.fragments.append( str(top_result.subject))
else:
gat_profile.fragments.append( str(top_result.subject)+ '\'')
gat_profile.orientate_for_dnaA()
# lookup the gat_profile to get the number
tg = TypeGenerator(p, gat_profile)
type_output_string = tg.calculate_type() + "\t" + str(gat_profile)
if not tg.has_previously_seen:
with open(self.novel_profiles, "a+") as output_fh:
output_fh.write(self.db_dir + "\t" + type_output_string + "\n")
return type_output_string
def test_populate_fragments_from_chromosome(self):
f = Fasta(os.path.join(data_dir, 'calc_fragment_coords.fa'), False)
fragments = f.calc_fragment_coords([[45, 55], [90, 110], [150, 180]])
sequences = [str(f.sequence) for f in fragments]
self.assertEqual(sequences, ["", "", ""])
f.populate_fragments_from_chromosome(fragments, None)
sequences = [str(f.sequence) for f in fragments]
self.assertEqual(sequences, [
"TTTTTTTTTTTTTTTTTTTTAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA",
"CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC",
"GGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGG"
])
def test_chop_from_ends(self):
f = Fasta(os.path.join(data_dir, 'calc_fragment_coords.fa'), False)
boundries = [
Operon(45, 55, True),
Operon(90, 110, False),
Operon(150, 180, True)
]
fragments = f.calc_fragment_coords(boundries)
sequences = [str(f.sequence) for f in fragments]
f.populate_fragments_from_chromosome(fragments, 5)
sequences = [str(f.sequence) for f in fragments]
self.assertEqual(sequences,
['TTTTTNNNAAAAA', 'CCCCCNNNCCCCC', 'GGGGGNNNGGGGG'])
def shrink_files(self):
output_filenames = []
# copy profile files to destination
shutil.copy(os.path.join(self.input_database, 'profile.txt'), self.output_database)
shutil.copy(os.path.join(self.input_database, 'profile.txt.yml'), self.output_database)
fasta_file_names_compressed = self.get_database_files_compressed()
fasta_file_names_uncompressed = self.get_database_files()
fasta_file_names = fasta_file_names_uncompressed + fasta_file_names_compressed
fasta_obj = [ Fasta(f) for f in fasta_file_names]
for f in fasta_obj:
fb = FilterBlast(self.blast_results, 1, 1)
destination_filename = os.path.join(self.output_database, str(f.fragment_number()) + '.fa')
if len(f.chromosome.seq) < self.target_bases:
if f.input_file in fasta_file_names_compressed:
shutil.copy(f.input_file, destination_filename + '.gz')
output_filenames.append(destination_filename + '.gz')
else:
shutil.copy(f.input_file, destination_filename)
output_filenames.append(destination_filename)
else:
blocks = fb.identify_regions(f.fragment_number(), self.target_bases)
sequence = ""
for b in blocks:
sequence += f.chromosome.seq[(b[0]):(b[1])]
record = [SeqRecord(sequence, str(f.fragment_number()) , '', '')]
SeqIO.write(record, destination_filename, "fasta")
output_filenames.append(destination_filename)
return self.compress_files(output_filenames)
def find_boundries(self, coords):
boundries = []
starting_coords = []
ending_coords = []
variable_s = self.five_or_23s(coords)
for c in coords:
if (c[2] == 16 and c[3] == '+') or (c[2] == variable_s
and c[3] == '-'):
# start of ribo
starting_coords.append(c[0])
elif (c[2] == 16 and c[3] == '-') or (c[2] == variable_s
and c[3] == '+'):
# end of ribo
ending_coords.append(c[1])
starting_coords = self.filter_out_close_start_coords(starting_coords)
ending_coords = self.filter_out_close_end_coords(ending_coords)
for start_index in range(len(starting_coords)):
start = starting_coords[start_index]
if start < 0:
continue
for end_index in range(len(ending_coords)):
end = ending_coords[end_index]
if end < 0:
continue
if end - start < self.len_70s and end - start > 0:
boundries.append([start, end])
ending_coords[end_index] = -1
starting_coords[start_index] = -1
continue
# check for 70S that goes over the end of the genome and for errors
remaining_start_coords = [s for s in starting_coords if s >= 0]
remaining_end_coords = [e for e in ending_coords if e >= 0]
if len(remaining_start_coords) > 0 and len(remaining_end_coords) > 0:
chromosome_length = self.chromosome_length
if self.chromosome_length <= 0:
chromosome_length = len(
Fasta(self.input_file, self.verbose).chromosome)
for start_index in range(len(remaining_start_coords)):
start = remaining_start_coords[start_index]
for end_index in range(len(remaining_end_coords)):
end = remaining_end_coords[end_index]
if end < 0:
continue
if (chromosome_length - start
) < self.len_70s and end < self.len_70s and (
chromosome_length - start) + end < self.len_70s:
boundries.append([start, end])
remaining_end_coords[end_index] = -1
remaining_start_coords[start_index] = -1
continue
return boundries
def test_populate_fragments_from_chromosome(self):
f = Fasta(os.path.join(data_dir, 'calc_fragment_coords.fa'), False)
boundries = [
Operon(45, 55, True),
Operon(90, 110, False),
Operon(150, 180, True)
]
fragments = f.calc_fragment_coords(boundries)
sequences = [str(f.sequence) for f in fragments]
self.assertEqual(sequences, ["", "", ""])
f.populate_fragments_from_chromosome(fragments, None)
sequences = [str(f.sequence) for f in fragments]
self.assertEqual(sequences, [
"TTTTTTTTTTTTTTTTTTTTAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA",
"CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC",
"GGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGG"
])
def test_fasta_get_largest_contig(self):
f = Fasta(os.path.join(data_dir, 'get_largest_contig.fa'), False)
largest_contig_record = f.get_chromosome_from_fasta()
self.assertEqual(len(largest_contig_record.seq), 60)
def populate_fragments_from_chromosome(self, input_file, boundries):
f = Fasta(input_file, self.verbose, is_circular=self.is_circular)
fragments = f.calc_fragment_coords(boundries)
f.populate_fragments_from_chromosome(fragments,
self.max_bases_from_ends)
return fragments