def run(self, manager):
mgra_ex_path = get_from_dict_with_path(manager.configuration,
key="executable_path",
path=["mgra"])
manager.logger.info("=" * 80)
if mgra_ex_path is None:
manager.logger.info(
"MGRA executable path is not supplied, skipping the MGRA based tasks"
)
return
manager.logger.info(
"Preparing data to communicate with MGRA and ontain guidance graph"
)
temp_dir = os.path.join(
manager.configuration["gos-asm"]["output"]["dir"], "tmp_mgra")
if not os.path.exists(temp_dir):
os.mkdir(temp_dir)
blocks_file_name = os.path.join(temp_dir, "blocks.txt")
config_file_name = os.path.join(temp_dir, "config.cfg")
mgra_output_dir_name = os.path.join(temp_dir, "output/")
manager.logger.debug(
"Writing blocks orders in GRIMM format to {file_name}".format(
file_name=blocks_file_name))
GRIMMWriter.print_genomes_as_grimm_blocks_orders(
bg=manager.data["gos-asm"]["bg"], file_name=blocks_file_name)
manager.logger.debug(
"Writing configuration file for MGRA run to {file_name}".format(
file_name=config_file_name))
config = self.create_mgra_config(blocks_file_name=blocks_file_name,
manager=manager)
with open(config_file_name, "wt") as destination:
json.dump(obj=config, fp=destination)
manager.logger.info("Running MGRA on prepared configuration")
os.system("{mgra_ex_path} -c {config_file_path} -o {output_dir_path}"
"".format(mgra_ex_path=mgra_ex_path,
config_file_path=config_file_name,
output_dir_path=mgra_output_dir_name))
manager.logger.debug("MGRA has successfully finished")
manager.logger.info("Reading MGRA produced guidance graph")
genomes_dir = os.path.join(mgra_output_dir_name, "genomes")
genome_files = [
name for name in os.listdir(genomes_dir) if name.endswith(".gen")
]
full_genomes_paths = [
os.path.join(genomes_dir, name) for name in genome_files
]
guidance_bg = BreakpointGraph()
for file_name in full_genomes_paths:
with open(file_name, "rt") as source:
guidance_bg.update(
breakpoint_graph=GRIMMReader.get_breakpoint_graph(
stream=source, merge_edges=False),
merge_edges=False)
if "mgra" not in manager.data:
manager.data["mgra"] = {}
manager.data["mgra"]["guidance_graph"] = guidance_bg
manager.logger.info("Obtained MGRA produced guidance graph")
def test_get_fragments_grimm_from_breakpoint_graph_single_genome_with_repeat_based_merges(
self):
data = [
">Mouse", "# data :: fragment : name = scaffold1",
"1 ALC__repeat $", "# data :: fragment : name = scaffold2",
"ALC__repeat 2 $", "# data :: fragment : name = scaffold3",
"ALC__repeat 3 $"
]
bg = self._populate_bg(data=data)
iv1 = bg.get_vertex_by_name("1h__repeat:ALCt__infinity")
iv2 = bg.get_vertex_by_name("2t__repeat:ALCh__infinity")
v1 = bg.get_vertex_by_name("1h")
v2 = bg.get_vertex_by_name("2t")
kbreak = KBreak(start_edges=[(v1, iv1), (v2, iv2)],
result_edges=[(v1, v2), (iv1, iv2)],
multicolor=Multicolor(BGGenome("Mouse")))
bg.apply_kbreak(kbreak=kbreak)
grimm_strings = GRIMMWriter.get_fragments_in_grimm_from_breakpoint_graph(
bg=bg)
possibilities_1 = ["scaffold1 scaffold2 $", "-scaffold2 -scaffold1 $"]
possibilities_3 = ["scaffold3 $", "-scaffold3 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_1)))
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_3)))
def test_get_grimm_from_breakpoint_graph_single_chromosome(self):
data = [">Mouse", "1 2 3 4 5 $"]
bg = self._populate_bg(data=data)
grimm_strings = GRIMMWriter.get_blocks_in_grimm_from_breakpoint_graph(
bg=bg)
self.assertEqual(len(grimm_strings), 2)
self.assertIn(">Mouse", grimm_strings)
possibilities_1 = ["1 2 3 4 5 $", "-5 -4 -3 -2 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_1)))
def test_get_grimm_strings_from_breakpoint_graph_single_genome(self):
self._populate_single_genome_bg()
grimm_strings = GRIMMWriter.get_blocks_in_grimm_from_breakpoint_graph(
bg=self.single_genome_bg)
self.assertEqual(len(grimm_strings), 3)
self.assertIn(">Mouse", grimm_strings)
possibilities_1 = ["1 2 3 4 $", "-4 -3 -2 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_1)))
possibilities_2 = ["5 6 7 8 $", "-8 -7 -6 -5 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_2)))
def test_output_genomes_as_grimm(self):
self._populate_four_genomes_bg(merge_edges=True)
file_name = "file_name.txt"
GRIMMWriter.print_genomes_as_grimm_blocks_orders(
bg=self.four_genome_bg, file_name=file_name)
try:
with open(file_name, "rt") as source:
new_bg = GRIMMReader.get_breakpoint_graph(stream=source,
merge_edges=True)
self.assertEqual(len(list(new_bg.nodes())),
len(list(self.four_genome_bg.nodes())))
self.assertEqual(len(list(new_bg.edges())),
len(list(self.four_genome_bg.edges())))
self.assertSetEqual(set(new_bg.nodes()),
set(self.four_genome_bg.nodes()))
self.assertSetEqual(
new_bg.get_overall_set_of_colors(),
self.four_genome_bg.get_overall_set_of_colors())
finally:
if os.path.exists(file_name):
os.remove(file_name)
def test_get_fragments_grimm_from_breakpoint_graph_single_genome(self):
data = [
">Mouse", "# data :: fragment : name = scaffold1",
"1 repeat__LC-1 $", "# data :: fragment : name = scaffold2", "2 $",
"# data :: fragment : name = scaffold3", "repeat__ALC 3 $"
]
bg = self._populate_bg(data=data)
grimm_strings = GRIMMWriter.get_fragments_in_grimm_from_breakpoint_graph(
bg=bg)
possibilities_1 = ["scaffold1 $", "-scaffold1 $"]
possibilities_2 = ["scaffold2 $", "-scaffold2 $"]
possibilities_3 = ["scaffold3 $", "-scaffold3 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_1)))
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_2)))
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_3)))
def test_get_grimm_strings_from_breakpoints_graph_two_genomes(self):
self._populate_two_genomes_bg()
grimm_strings = GRIMMWriter.get_blocks_in_grimm_from_breakpoint_graph(
bg=self.two_genome_bg)
self.assertEqual(len(grimm_strings), 6)
self.assertIn(">Mouse", grimm_strings)
self.assertIn(">Human", grimm_strings)
possibilities_1 = ["1 2 3 4 $", "-4 -3 -2 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_1)))
possibilities_2 = ["5 6 7 8 $", "-8 -7 -6 -5 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_2)))
possibilities_3 = ["1 4 3 2 $", "-2 -3 -4 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_3)))
possibilities_4 = ["5 -7 -6 8 $", "-8 6 7 -5 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_4)))
def test_get_grimm_from_breakpoint_graph_four_genomes(self):
self._populate_four_genomes_bg()
grimm_strings = GRIMMWriter.get_blocks_in_grimm_from_breakpoint_graph(
bg=self.four_genome_bg)
self.assertEqual(len(grimm_strings), 12)
self.assertIn(">Mouse", grimm_strings)
self.assertIn(">Human", grimm_strings)
self.assertIn(">Rat", grimm_strings)
self.assertIn(">Chimpanzee", grimm_strings)
possibilities_1 = ["1 2 3 4 $", "-4 -3 -2 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_1)))
possibilities_2 = ["5 6 7 8 $", "-8 -7 -6 -5 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_2)))
possibilities_3 = ["1 4 3 2 $", "-2 -3 -4 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_3)))
possibilities_4 = ["5 -7 -6 8 $", "-8 6 7 -5 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_4)))
possibilities_5 = ["0 1 4 5 $", "-5 -4 -1 -0 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_5)))
possibilities_6 = ["10 12 8 7 $", "-7 -8 -12 -10 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_6)))
possibilities_7 = ["5 6 7 8 $", "-8 -7 -6 -5 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_7)))
possibilities_8 = ["1 -4 -3 -2 $", "2 3 4 -1 $"]
self.assertTrue(
any(map(lambda entry: entry in grimm_strings, possibilities_8)))