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 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 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