def main():
fastg_file_name = ''
blast_result_file = ''
output_file = ''
overlap_len = None
options, args = getopt.getopt(sys.argv[1:], 'hl:')
for option, value in options:
if option == '-l':
overlap_len = int(value)
elif option == '-h':
print_help_message()
sys.exit()
else:
print_help_message()
sys.exit()
fastg_file_name, blast_result_file, output_file = args
nodes = fastg_file.build_assembly_graph(fastg_file_name,
overlap_len)
alignments = list(filter(lambda x: x.is_valid and x.is_forward,
read_file(blast_result_file)))
Alignment.add_connection(alignments, nodes)
alignments.sort(key=lambda x: x.start)
# write_file(output_file, alignments)
values, actions = Alignment.get_path(alignments)
# Alignment.write_alignments_to_dot_file(alignments, output_file,
# actions, values)
Alignment.write_path_to_dot_file(actions, values, output_file)
def _test_read_file():
input_file = 'assembly_graph.siteGraph'
fastg_file_ = 'assembly_graph.fastg'
import fastg_file
nodes = fastg_file.build_assembly_graph(fastg_file_, 127)
sites = read_file(input_file, nodes)
for site in sites.values():
print(site.id)
print([(ele[0], ele[1]) for ele in site.children])
def main():
input_file = ''
output_file = ''
overlap_len = None
shift_len = 0
to_simplify = False
max_interval_len = None
options, args = getopt.getopt(sys.argv[1:], 'm:k:i:l:o:hs')
for option, value in options:
if option == '-i':
input_file = value
elif option == '-o':
output_file = value
elif option == '-l':
overlap_len = int(value)
elif option == '-k':
shift = int(value)
elif option == '-m':
max_interval_len = int(value)
elif option == '-s':
to_simplify = True
elif option == '-h':
printHelpMessage()
sys.exit()
else:
printHelpMessage()
sys.exit()
print('overlap_len:', overlap_len)
nodes = fastg_file.build_assembly_graph(input_file, overlap=overlap_len)
tick = time.time()
sites, site_position_index = build_site_graph(nodes, shift=shift)
tock = time.time()
if to_simplify:
print('Simplifying site graph...')
sites = site_graph.simplify_site_graph(sites)
# print('{} sites created on {} nodes'.format(len(sites), len(site_position_index)))
tock2 = time.time()
# debug_site = sites['648r']
# print('site', debug_site.id, 'has {} children.'.format(len(debug_site.children)))
# for child, interval, nodes_path, _ in debug_site.children:
# print(child.id, interval, [ele.uid for ele in nodes_path])
site_graph.write_file(output_file, sites, [
' '.join(sys.argv), 'Number of sites: {}'.format(
len(sites)), 'Number of nodes: {}'.format(len(nodes)),
'Numner of nodes contain site: {}'.format(len(site_position_index)),
'Max interval len: {}'.format(max_interval_len),
'Allowed repeat num: {}'.format(ALLOWED_REPEAT_NUM),
'Time used to build graph: {} seconds'.format(round(tock - tick)),
'Time used to simplify graph: {} seconds'.format(round(tock2 - tock))
])
def main():
interface = 'hk:'
options, args = getopt.getopt(sys.argv[1:], interface)
overlap_size = None
for option, value in options:
if option == '-h':
print_help()
sys.exit()
elif option == '-k':
overlap_size = int(value)
input_file, output_file = args
nodes = fastg_file.build_assembly_graph(input_file, overlap_size)
write_last_graph(nodes, output_file, overlap_size)
def _test_build_site_graph():
input_file = 'assembly_graph.fastg'
# input_file = 'test.fastg'
overlap = 127
nodes = fastg_file.build_assembly_graph(input_file, overlap)
# nodes = fastg_file.build_assembly_graph(input_file, overlap=2)
sites, site_position_index = build_site_graph(nodes)
num_position_in_index = sum(
(len(ele) for ele in site_position_index.values()))
site_ids = [
int(site.id) for site in sites.values() if not site.id.endswith('r')
]
site_ids.sort()
site_positions = list(site_position_index.items())
site_positions.sort(key=lambda x: int(x[0].uid.rstrip('r')))
site_index = {}
count = [0] * 10
num_site_contain_self_as_child = 0
for node, positions in site_position_index.items():
for position, site in positions.items():
if site not in site_index:
site_index[site] = [(node.uid, position)]
else:
site_index[site].append((node.uid, position))
for v in site_index.values():
v.sort(key=lambda x: int(x[0].rstrip('r')))
site_index_items = list(site_index.items())
site_index_items.sort(key=lambda x: x[1][0])
for site, positions in site_index_items:
child_sites, intervals, node_paths = list(zip(
*site.children)) if site.children else ([], [], [])
child_site_ids = [ele.id for ele in child_sites]
node_ids = [[ele.uid for ele in node_path] for node_path in node_paths]
child_site_id_intervals = list(zip(child_site_ids, intervals))
child_site_id_intervals_ = sorted(child_site_id_intervals)
# print(site, positions, 'C:', child_site_id_intervals_)
if site.id in child_site_ids:
num_site_contain_self_as_child += 1
print('!!!')
count[len(positions)] += 1
def main():
paths_file_name = ''
gap_info_file_name = ''
fastg_file_name = ''
interface = 'hp:g:'
options, args = getopt.getopt(sys.argv[1:], interface)
for option, value in options:
if option == '-h':
print_help()
sys.exit()
elif option == '-p':
paths_file_name = value
elif option == '-g':
fastg_file_name = value
gap_info_file_name, output_file_name = args
# paths_file_name = '/home/huangbin/simulation_lab/Bacteria/E.coli_PL/spades.dir/contigs.paths'
# gap_info_file_name = '/home/huangbin/simulation_lab/Bacteria/E.coli_PL/hybrid_scaffold_keep_11.dir/hybrid_scaffolds/tmp1.txt'
# fastg_file_name = 'datasets/assembly_graph.fastg'
paths = paths_file.read_file(paths_file_name)
gaps = gap_info_file.read_file(gap_info_file_name)
nodes = fastg_file.build_assembly_graph(fastg_file_name, overlap=OVERLAP)
_, site_position_indexs = sitegraph_builder.build_site_graph(nodes, mode=1)
# Transform gaps.
for gap in gaps:
debug = 0
if gap.start_node_id.startswith('NODE_2_'):
debug = 1
if debug == 1:
print('DEBUG:', gap.start_node_id, gap.start_site_index)
gap.start_node_id, gap.start_site_index, gap.start_site_position =\
transform_position(gap.start_node_id, gap.start_site_position, nodes, paths, site_position_indexs, OVERLAP, debug=debug)
gap.end_node_id, gap.end_site_index, gap.end_site_position =\
transform_position(gap.end_node_id, gap.end_site_position, nodes, paths, site_position_indexs, OVERLAP, debug=debug)
if debug == 1:
print('DEBUG:', gap.start_node_id, gap.start_site_index)
# Write gaps.
cmd = ' '.join(sys.argv)
gap_info_file.write_file(gaps, output_file_name, comments=[cmd])
def main():
# unmodified_reference_cmap = None
global find_path_script
fastg_file_name = None
site_graph_file_name = None
find_path_script = 'find_path_dp.py'
work_dir = None
task_name = None
n_thread = 1
is_node_id_processed = 0
interface = 'hf:o:n:x:s:m:'
options, args = getopt.getopt(sys.argv[1:], interface)
for option, value in options:
if option == '-h':
print_help()
sys.exit()
elif option == '-f':
fastg_file_name = value
elif option == '-s':
site_graph_file_name = value
elif option == '-x':
find_path_script = value
elif option == '-n':
n_thread = int(value)
elif option == '-o':
overlap_len = int(value)
elif option == '-m':
is_node_id_processed = int(value)
gap_info_file_name = args[0]
if len(args) == 1:
pass
if len(args) == 2:
task_name = args[1]
elif len(args) == 3:
work_dir, task_name = args[1:3]
else:
print_help()
sys.exit()
if task_name is None:
task_name = fastg_file_name.rsplit('.', 1)
global log_dir, seq_fa_file, graph_pickle_file
log_dir = work_dir + '/' + task_name + '_log.dir'
seq_fa_file = work_dir + '/' + task_name + '.fa'
graph_pickle_file = work_dir + '/' + task_name + '_graph.pickle'
subprocess.run(('rm', '-rf', log_dir))
subprocess.run(('mkdir', log_dir))
# Read assembly graph.
nodes = fastg_file.build_assembly_graph(fastg_file_name, overlap_len)
if site_graph_file_name:
# Reads site graph from file.
sites = site_graph.read_file(site_graph_file_name, nodes)
_, site_position_index = sitegraph_builder.build_site_graph(nodes,
mode=1)
for position_index in site_position_index.values():
for position in position_index:
original_site = position_index[position]
position_index[position] = sites[original_site.id]
else:
# Build site graph form assembly graph, and write site graph.
sites, site_position_index = \
sitegraph_builder.build_site_graph(nodes)
sites = site_graph.simplify_site_graph(sites)
site_graph_file_name = fastg_file_name.rsplit('.')[0] +\
'.sitegraph'
site_graph.write_file(site_graph_file_name, sites)
# Write nodes and sites to pickle file.
with open(graph_pickle_file, 'wb') as fout:
pickle.dump(nodes, fout, -1)
pickle.dump(sites, fout, -1)
# Find start site and end site.
gaps = gap_info_file.read_file(gap_info_file_name)
exe_tuples = []
for gap in gaps:
if is_node_id_processed:
start_node_id = gap.start_node_id
end_node_id = gap.end_node_id
else:
start_node_id = get_node_id_from_long_name(gap.start_node_id)
end_node_id = get_node_id_from_long_name(gap.end_node_id)
start_site_position, start_site = get_site_by_index(
gap.start_site_index, site_position_index[nodes[start_node_id]])
end_site_position, end_site = get_site_by_index(
gap.end_site_index, site_position_index[nodes[end_node_id]])
print(start_site.id, end_site.id)
gap_seq_id = '-'.join((start_node_id, end_node_id))
exe_tuples.append(
(gap_seq_id,
construct_cmd(start_site.id, end_site.id, start_site_position,
end_site_position, graph_pickle_file, gap.intervals,
gap_seq_id)))
with Pool(n_thread) as p:
p.map(find_path_dp_process, exe_tuples)