def main():
time0 = time.time()
print_title = "GetOrganelle v" + str(get_versions()) + \
"\n\nThis is a script for extracting organelle genomes" \
" from slim_fastg.py-produced files (csv & fastg). " + \
"\nBy [email protected]\n\n"
options, log_handler = get_options(print_title)
@set_time_limit(options.time_limit)
def disentangle_circular_assembly(fastg_file,
tab_file,
prefix,
weight_factor,
type_factor,
mode="embplant_pt",
log_hard_cov_threshold=10.,
expected_max_size=inf,
expected_min_size=0,
contamination_depth=3.,
contamination_similarity=5.,
degenerate=True,
degenerate_depth=1.5,
degenerate_similarity=1.5,
min_sigma_factor=0.1,
max_copy_in=10,
only_max_cov=True,
keep_temp=False,
acyclic_allowed=False,
verbose=False,
inner_logging=None,
debug=False):
if options.resume and os.path.exists(prefix +
".graph1.selected_graph.gfa"):
pass
if inner_logging:
inner_logging.info(">>> Result graph existed!")
else:
sys.stdout.write(">>> Result graph existed!\n")
else:
time_a = time.time()
if inner_logging:
inner_logging.info(">>> Parsing " + fastg_file + " ..")
else:
sys.stdout.write("Parsing " + fastg_file + " ..\n")
input_graph = Assembly(fastg_file,
min_cov=options.min_cov,
max_cov=options.max_cov)
time_b = time.time()
if inner_logging:
inner_logging.info(">>> Parsing input fastg file finished: " +
str(round(time_b - time_a, 4)) + "s")
else:
sys.stdout.write("\n>>> Parsing input fastg file finished: " +
str(round(time_b - time_a, 4)) + "s\n")
temp_graph = prefix + ".temp.fastg" if keep_temp else None
copy_results = input_graph.find_target_graph(
tab_file,
database_name=mode,
mode=mode,
type_factor=type_factor,
weight_factor=weight_factor,
log_hard_cov_threshold=log_hard_cov_threshold,
contamination_depth=contamination_depth,
contamination_similarity=contamination_similarity,
degenerate=degenerate,
degenerate_depth=degenerate_depth,
degenerate_similarity=degenerate_similarity,
expected_max_size=expected_max_size,
expected_min_size=expected_min_size,
max_contig_multiplicity=max_copy_in,
only_keep_max_cov=only_max_cov,
min_sigma_factor=min_sigma_factor,
temp_graph=temp_graph,
broken_graph_allowed=acyclic_allowed,
verbose=verbose,
log_handler=inner_logging,
debug=debug)
time_c = time.time()
if inner_logging:
inner_logging.info(">>> Detecting target graph finished: " +
str(round(time_c - time_b, 4)) + "s")
if len(copy_results) > 1:
inner_logging.info(
str(len(copy_results)) + " set(s) of graph detected.")
else:
sys.stdout.write("\n\n>>> Detecting target graph finished: " +
str(round(time_c - time_b, 4)) + "s\n")
if len(copy_results) > 1:
sys.stdout.write(
str(len(copy_results)) +
" set(s) of graph detected.\n")
degenerate_base_used = False
if acyclic_allowed:
# still_complete = []
for go_res, copy_res in enumerate(copy_results):
go_res += 1
broken_graph = copy_res["graph"]
count_path = 0
these_paths = broken_graph.get_all_paths(
mode=mode, log_handler=inner_logging)
# reducing paths
if len(these_paths) > options.max_paths_num:
this_warn_str = "Only exporting " + str(options.max_paths_num) + " out of all " + \
str(len(these_paths)) + " possible paths. (see '--max-paths-num' to change it.)"
if inner_logging:
inner_logging.warning(this_warn_str)
else:
sys.stdout.write("Warning: " + this_warn_str +
"\n")
these_paths = these_paths[:options.max_paths_num]
# exporting paths, reporting results
for this_paths, other_tag in these_paths:
count_path += 1
all_contig_str = []
contigs_are_circular = []
for go_contig, this_p_part in enumerate(this_paths):
this_contig = broken_graph.export_path(this_p_part)
if DEGENERATE_BASES & set(this_contig.seq):
degenerate_base_used = True
if this_contig.label.endswith("(circular)"):
contigs_are_circular.append(True)
else:
contigs_are_circular.append(False)
if len(this_paths
) == 1 and contigs_are_circular[-1]:
all_contig_str.append(this_contig.fasta_str())
else:
all_contig_str.append(">contig_" +
str(go_contig + 1) +
"--" +
this_contig.label +
"\n" + this_contig.seq +
"\n")
if len(all_contig_str) == 1 and set(
contigs_are_circular) == {True}:
# print ir stat
if count_path == 1 and mode == "embplant_pt":
detect_seq = broken_graph.export_path(
this_paths[0]).seq
ir_stats = detect_plastome_architecture(
detect_seq, 1000)
print_str = "Detecting large repeats (>1000 bp) in PATH1 with " + ir_stats[-1] +\
", Total:LSC:SSC:Repeat(bp) = " + str(len(detect_seq)) + ":" + \
":".join([str(len_val) for len_val in ir_stats[:3]])
if inner_logging:
inner_logging.info(print_str)
else:
sys.stdout.write(print_str + "\n")
# if len(all_contig_str) == 1 and set(contigs_are_circular) == {True}:
# still_complete.append(True)
# else:
# still_complete.append(False)
open(
prefix + ".graph" + str(go_res) + other_tag + "." +
str(count_path) + ".path_sequence.fasta",
"w").write("\n".join(all_contig_str))
broken_graph.write_to_gfa(prefix + ".graph" + str(go_res) +
".selected_graph.gfa")
else:
for go_res, copy_res in enumerate(copy_results):
go_res += 1
idealized_graph = copy_res["graph"]
# should add making one-step-inversion pairs for paths,
# which would be used to identify existence of a certain isomer using mapping information
count_path = 0
these_paths = idealized_graph.get_all_circular_paths(
mode=mode,
log_handler=inner_logging,
reverse_start_direction_for_pt=options.reverse_lsc)
# reducing paths
if len(these_paths) > options.max_paths_num:
this_warn_str = "Only exporting " + str(options.max_paths_num) + " out of all " + \
str(len(these_paths)) + " possible paths. (see '--max-paths-num' to change it.)"
if inner_logging:
inner_logging.warning(this_warn_str)
else:
sys.stdout.write("Warning: " + this_warn_str +
"\n")
these_paths = these_paths[:options.max_paths_num]
# exporting paths, reporting results
for this_path, other_tag in these_paths:
count_path += 1
this_seq_obj = idealized_graph.export_path(this_path)
if DEGENERATE_BASES & set(this_seq_obj.seq):
degenerate_base_used = True
open(
prefix + ".graph" + str(go_res) + other_tag + "." +
str(count_path) + ".path_sequence.fasta",
"w").write(this_seq_obj.fasta_str())
# print ir stat
if count_path == 1 and mode == "embplant_pt":
detect_seq = this_seq_obj.seq
ir_stats = detect_plastome_architecture(
detect_seq, 1000)
print_str = "Detecting large repeats (>1000 bp) in PATH1 with " + ir_stats[-1] + \
", Total:LSC:SSC:Repeat(bp) = " + str(len(detect_seq)) + ":" + \
":".join([str(len_val) for len_val in ir_stats[:3]])
if inner_logging:
inner_logging.info(print_str)
else:
sys.stdout.write(print_str + "\n")
idealized_graph.write_to_gfa(prefix + ".graph" +
str(go_res) +
".selected_graph.gfa")
if degenerate_base_used:
inner_logging.warning("Degenerate base(s) used!")
time_d = time.time()
if inner_logging:
inner_logging.info(
">>> Solving and unfolding graph finished: " +
str(round(time_d - time_c, 4)) + "s")
else:
sys.stdout.write(
"\n\n>>> Solving and unfolding graph finished: " +
str(round(time_d - time_c, 4)) + "s\n")
try:
disentangle_circular_assembly(
options.fastg_file,
options.tab_file,
os.path.join(options.output_directory, options.prefix),
type_factor=options.type_factor,
mode=options.mode,
weight_factor=options.weight_factor,
log_hard_cov_threshold=options.depth_factor,
contamination_depth=options.contamination_depth,
contamination_similarity=options.contamination_similarity,
degenerate=options.degenerate,
degenerate_depth=options.degenerate_depth,
degenerate_similarity=options.degenerate_similarity,
expected_max_size=options.expected_max_size,
expected_min_size=options.expected_min_size,
min_sigma_factor=options.min_sigma_factor,
max_copy_in=options.max_multiplicity,
only_max_cov=options.only_keep_max_cov,
acyclic_allowed=options.acyclic_allowed,
keep_temp=options.keep_temp_graph,
inner_logging=log_handler,
verbose=options.verbose,
debug=options.debug)
log_handler = simple_log(logging.getLogger(), options.output_directory,
options.prefix + ".disentangle.")
log_handler.info('\nTotal cost: ' +
str(round(time.time() - time0, 4)) + 's\n')
except IOError as e:
raise e
except KeyError as e:
if str(e).strip("'") == options.mode:
log_handler.error(options.mode + " not found in " +
str(options.tab_file) + "!")
log_handler.error("Disentangling failed!")
else:
log_handler.exception(str(e))
log_handler.error("Disentangling failed!")
if not options.acyclic_allowed:
log_handler.info(
"You might try again with '--linear' to export contig(s) "
"instead of circular genome.")
log_handler = simple_log(log_handler, options.output_directory,
options.prefix + ".disentangle.")
log_handler.info("\nTotal cost " + str(time.time() - time0))
log_handler.info(
"Please email [email protected] if you find bugs!\n")
except Exception as e:
log_handler.exception(str(e))
log_handler.error("Disentangling failed!")
if not options.acyclic_allowed:
log_handler.info(
"You might try again with '--linear' to export contig(s) "
"instead of circular genome.")
log_handler = simple_log(log_handler, options.output_directory,
options.prefix + ".disentangle.")
log_handler.info("\nTotal cost " + str(time.time() - time0))
log_handler.info(
"Please email [email protected] if you find bugs!\n")
logging.shutdown()
def main():
time0 = time.time()
print_title = "GetOrganelle v" + str(get_versions()) + \
"\n\nThis is a script for extracting organelle genomes" \
" from slim_fastg.py-produced files (csv & fastg). " + \
"\nBy [email protected]\n\n"
options, log_handler = get_options(print_title)
@set_time_limit(options.time_limit)
def disentangle_circular_assembly(fastg_file, tab_file, prefix, weight_factor, type_factor, mode="embplant_pt",
log_hard_cov_threshold=10., expected_max_size=inf, expected_min_size=0,
contamination_depth=3., contamination_similarity=5.,
degenerate=True, degenerate_depth=1.5, degenerate_similarity=1.5,
min_sigma_factor=0.1, only_max_c=True, keep_temp=False, acyclic_allowed=False,
verbose=False, log_handler=None, debug=False):
if options.resume and os.path.exists(prefix + ".graph1.selected_graph.gfa"):
pass
if log_handler:
log_handler.info(">>> Result graph existed!")
else:
sys.stdout.write(">>> Result graph existed!\n")
else:
time_a = time.time()
if log_handler:
log_handler.info(">>> Parsing " + fastg_file + " ..")
else:
sys.stdout.write("Parsing " + fastg_file + " ..\n")
input_graph = Assembly(fastg_file, min_cov=options.min_cov, max_cov=options.max_cov)
time_b = time.time()
if log_handler:
log_handler.info(">>> Parsing input fastg file finished: " + str(round(time_b - time_a, 4)) + "s")
else:
sys.stdout.write("\n>>> Parsing input fastg file finished: " + str(round(time_b - time_a, 4)) + "s\n")
temp_graph = prefix + ".temp.fastg" if keep_temp else None
copy_results = input_graph.find_target_graph(tab_file, mode=mode, type_factor=type_factor,
weight_factor=weight_factor,
log_hard_cov_threshold=log_hard_cov_threshold,
contamination_depth=contamination_depth,
contamination_similarity=contamination_similarity,
degenerate=degenerate, degenerate_depth=degenerate_depth,
degenerate_similarity=degenerate_similarity,
expected_max_size=expected_max_size,
expected_min_size=expected_min_size,
only_keep_max_cov=only_max_c,
min_sigma_factor=min_sigma_factor,
temp_graph=temp_graph,
broken_graph_allowed=acyclic_allowed,
verbose=verbose, log_handler=log_handler,
debug=debug)
time_c = time.time()
if log_handler:
log_handler.info(">>> Detecting target graph finished: " + str(round(time_c - time_b, 4)) + "s")
if len(copy_results) > 1:
log_handler.info(str(len(copy_results)) + " set(s) of graph detected.")
else:
sys.stdout.write("\n\n>>> Detecting target graph finished: " + str(round(time_c - time_b, 4)) + "s\n")
if len(copy_results) > 1:
sys.stdout.write(str(len(copy_results)) + " set(s) of graph detected.\n")
degenerate_base_used = False
if acyclic_allowed:
# still_complete = []
for go_res, copy_res in enumerate(copy_results):
broken_graph = copy_res["graph"]
count_path = 0
for this_paths, other_tag in broken_graph.get_all_paths(mode=mode, log_handler=log_handler):
count_path += 1
all_contig_str = []
contigs_are_circular = []
for go_contig, this_p_part in enumerate(this_paths):
this_contig = broken_graph.export_path(this_p_part)
if DEGENERATE_BASES & set(this_contig.seq):
degenerate_base_used = True
if this_contig.label.endswith("(circular)"):
contigs_are_circular.append(True)
else:
contigs_are_circular.append(False)
if len(this_paths) == 1 and contigs_are_circular[-1]:
all_contig_str.append(this_contig.fasta_str())
else:
all_contig_str.append(">contig_" + str(go_contig + 1) + "--" + this_contig.label +
"\n" + this_contig.seq + "\n")
# if len(all_contig_str) == 1 and set(contigs_are_circular) == {True}:
# still_complete.append(True)
# else:
# still_complete.append(False)
open(prefix + ".graph" + str(go_res + 1) + other_tag + "." + str(count_path) +
".path_sequence.fasta", "w").write("\n".join(all_contig_str))
broken_graph.write_to_gfa(prefix + ".graph" + str(go_res + 1) + ".selected_graph.gfa")
else:
for go_res, copy_res in enumerate(copy_results):
idealized_graph = copy_res["graph"]
# should add making one-step-inversion pairs for paths,
# which would be used to identify existence of a certain isomer using mapping information
count_path = 0
for this_path, other_tag in idealized_graph.get_all_circular_paths(mode=mode, log_handler=log_handler):
count_path += 1
this_seq_obj = idealized_graph.export_path(this_path)
if DEGENERATE_BASES & set(this_seq_obj.seq):
degenerate_base_used = True
open(prefix + ".graph" + str(go_res + 1) + other_tag + "." + str(count_path) +
".path_sequence.fasta", "w").write(this_seq_obj.fasta_str())
idealized_graph.write_to_gfa(prefix + ".graph" + str(go_res + 1) + ".selected_graph.gfa")
if degenerate_base_used:
log_handler.warning("Degenerate base(s) used!")
time_d = time.time()
if log_handler:
log_handler.info(">>> Solving and unfolding graph finished: " + str(round(time_d - time_c, 4)) + "s")
else:
sys.stdout.write("\n\n>>> Solving and unfolding graph finished: " + str(round(time_d - time_c, 4)) + "s\n")
try:
disentangle_circular_assembly(options.fastg_file, options.tab_file,
os.path.join(options.output_directory, options.prefix),
type_factor=options.type_factor,
mode=options.mode,
weight_factor=options.weight_factor,
log_hard_cov_threshold=options.depth_factor,
contamination_depth=options.contamination_depth,
contamination_similarity=options.contamination_similarity,
degenerate=options.degenerate, degenerate_depth=options.degenerate_depth,
degenerate_similarity=options.degenerate_similarity,
expected_max_size=options.expected_max_size,
expected_min_size=options.expected_min_size,
min_sigma_factor=options.min_sigma_factor,
only_max_c=options.only_keep_max_cov, acyclic_allowed=options.acyclic_allowed,
keep_temp=options.keep_temp_graph,
log_handler=log_handler, verbose=options.verbose, debug=options.debug)
log_handler = simple_log(logging.getLogger(), options.output_directory, options.prefix + ".disentangle.")
log_handler.info('\nTotal cost: ' + str(round(time.time() - time0, 4)) + 's\n')
except Exception as e:
if options.debug:
log_handler.exception("")
else:
log_handler.exception(str(e))
log_handler.exception("Disentangling failed!")
if not options.acyclic_allowed:
log_handler.info("You might try again with '--linear' to export contig(s) instead of circular genome.")
log_handler = simple_log(log_handler, options.output_directory, options.prefix + ".disentangle.")
log_handler.info("\nTotal cost " + str(time.time() - time0))
log_handler.info("Please email [email protected] if you find bugs!\n")
logging.shutdown()