def clear_product_blocks(blocks):
blocks_samples = list(blocks.keys())
blocks_samples.sort(reverse=True)
iter = 0
total_blocks = make_block_stats(blocks)["total"]
for samples_num in blocks_samples:
patterns_numbers = list(blocks[samples_num].keys())
patterns_numbers.sort(reverse=True)
for patterns_num in patterns_numbers:
good_blocks = []
for block in blocks[samples_num][patterns_num]:
iter += 1
if iter % LOG_ITERATION == 0:
print("Product blocks checked {}/{}".format(
iter, total_blocks))
block_good = True
prod_finder = ProductFinder(blocks,
samples_num,
patterns_num,
equal=True)
products = prod_finder.find_products()
for product in products:
if len(product.product_list) == 1:
continue
if not block_good:
break
for product_block in product:
if product_block == block:
block_good = False
break
if block_good:
good_blocks.append(block)
blocks[samples_num][patterns_num] = good_blocks
return clear_empty_block_types(blocks)
def main():
args = read_args()
ncs, msg = find_ncs(args.ncs, script_path)
if not ncs:
print(msg)
exit()
result, blocks, ncs_name, deuterated = read_blocks(args.elb_file)
if result:
exit()
product_finder = ProductFinder(blocks, args.samples, args.min_patterns)
products = product_finder.find_products()
schemes_total = sum([len(product) for product in products])
blocks_total = {}
print("{} products found".format(schemes_total))
output = "[NCS = {}]\n".format(ncs_name)
output += "[Deuterated = {}]\n".format(deuterated)
for product in products:
for scheme in product:
scheme.simplify()
blocks_total = add_block(blocks_total, scheme.samples,
len(scheme.patterns))
output += scheme.full_str()
output += "\n"
stats_total = make_block_stats(blocks_total)
if args.output:
with open(args.output, mode="w") as f:
f.write(output)
print("{} blocks were written to the file \'{}\'".format(
schemes_total, args.output))
else:
print(output)
print("Statistics of generated blocks:")
print(stats_total["str"])
def clear_redundant_blocks(blocks):
iter = 0
total_blocks = make_block_stats(blocks)["total"]
for samples_num in blocks:
patterns_numbers = list(blocks[samples_num].keys())
patterns_numbers.sort(reverse=True)
good_blocks = []
for pattern_num in patterns_numbers:
new_block_list = []
for block in blocks[samples_num][pattern_num]:
iter += 1
if iter % LOG_ITERATION == 0:
print("Redundant blocks checked {}/{}".format(
iter, total_blocks))
block_good = True
for good_block in good_blocks:
if block.is_subset_of(good_block.simplified):
block_good = False
break
if block_good:
good_blocks.append(block)
new_block_list.append(block)
if not new_block_list:
blocks[samples_num].pop(pattern_num)
else:
blocks[samples_num][pattern_num] = new_block_list
return clear_empty_block_types(blocks)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("files",
help='Specify file(s) for which to calculate stats',
type=str,
nargs='+')
parser.add_argument("--total",
"-t",
help='Print only total number of blocks',
action="store_true")
args = parser.parse_args()
blocks_total = {}
for file in args.files:
blocks = {}
if not os.path.isfile(file):
print("Error! File '{}' not found!".format(file))
return
output = "{:>25}:".format(file)
with open(file, "r") as f:
for line in f:
result = Constants.elb_re.match(line)
if result:
blocks = add_block(blocks, int(result.group(1)),
int(result.group(2)))
blocks_total = add_block(blocks_total,
int(result.group(1)),
int(result.group(2)))
stats = make_block_stats(blocks)
if args.total:
output += " {:>7}".format(int(stats["total"]))
else:
output += "\n" + stats["str"]
print(output)
stats_total = make_block_stats(blocks_total)
if len(args.files) > 1:
output = "{:>25}:".format("TOTAL FOR {} FILES".format(len(args.files)))
if args.total:
output += " {:>7}".format(int(stats_total["total"]))
else:
output += "\n" + stats_total["str"]
print(output)
def clear_blocks(blocks, flags):
(identical_flag, redundant_flag, product_flag) = flags
print(make_block_stats(blocks)["str"] + "\n")
if identical_flag:
print("Clearing identical blocks...")
blocks = clear_identical_blocks(blocks)
print("Identical blocks cleared...")
print(make_block_stats(blocks)["str"] + "\n")
if redundant_flag:
print("Clearing redundant blocks...")
blocks = clear_redundant_blocks(blocks)
print("Redundant blocks cleared...")
print(make_block_stats(blocks)["str"] + "\n")
if product_flag:
print("Clearing product blocks...")
blocks = clear_product_blocks(blocks)
print("Product blocks cleared...")
print(make_block_stats(blocks)["str"] + "\n")
return blocks
def clear_identical_blocks(blocks):
iter = 0
total_blocks = make_block_stats(blocks)["total"]
for samples_num in blocks:
patterns_numbers = list(blocks[samples_num].keys())
patterns_numbers.sort(reverse=True)
for pattern_num in patterns_numbers:
new_block_dict = {}
for block_one in blocks[samples_num][pattern_num]:
iter += 1
if iter % LOG_ITERATION == 0:
print("Identical blocks checked {}/{}".format(
iter, total_blocks))
new_block_dict[block_one] = block_one
if not new_block_dict:
blocks[samples_num].pop(pattern_num)
else:
blocks[samples_num][pattern_num] = list(
new_block_dict.values())
return clear_empty_block_types(blocks)
def filterT_blocks(blocks, min_t_free, verbose_output):
iter = 0
total_blocks = make_block_stats(blocks)["total"]
filtered_blocks = 0
print(
"Filter blocks with less then {} patterns that have no \"T\" labeling".
format(min_t_free))
for samples_num in blocks:
patterns_numbers = list(blocks[samples_num].keys())
patterns_numbers.sort(reverse=True)
for pattern_num in patterns_numbers:
new_block_list = []
for block in blocks[samples_num][pattern_num]:
iter += 1
if iter % LOG_ITERATION == 0:
print("filterT blocks checked {}/{}".format(
iter, total_blocks))
have_t, have_no_t = count_typeT(block)
output = "have_t = {:2} have_no_t = {:2}, ".format(
have_t, have_no_t)
if have_no_t >= min_t_free:
output += "APPEND block"
new_block_list.append(block)
filtered_blocks += 1
else:
output += "SKIP block"
if verbose_output:
print(block.full_str())
print(output + "\n")
if not new_block_list:
blocks[samples_num].pop(pattern_num)
else:
blocks[samples_num][pattern_num] = new_block_list
print("Filtered {}/{} blocks by T labeling, min_t_free = {}".format(
filtered_blocks, total_blocks, min_t_free))
return clear_empty_block_types(blocks)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("files",
help='Specify file(s) for which to calculate stats',
type=str,
nargs='+')
parser.add_argument("--total",
"-t",
help='Print only total number of blocks',
action="store_true")
parser.add_argument("--sv",
"-s",
help='Print SV min andd max values',
action="store_true")
args = parser.parse_args()
blocks_total = {}
sv_min = dict()
sv_max = dict()
for file in args.files:
blocks = {}
if not os.path.isfile(file):
print("Error! File '{}' not found!".format(file))
return
output = "{:>25}:".format(file)
with open(file, "r") as f:
for line in f:
result = Constants.elb_re.match(line)
if result:
samples = int(result.group('samples'))
patterns = int(result.group('patterns'))
#print("{:>2} {:>2}".format(samples, patterns))
blocks = add_block(blocks, samples, patterns)
blocks_total = add_block(blocks_total, samples, patterns)
continue
if args.sv:
sv_re_match = Constants.sv_re.match(line)
if sv_re_match:
sv = [int(i) for i in sv_re_match.group('sv').split()]
if samples in sv_min.keys():
sv_min[samples] = list(
map(min, zip(sv_min[samples], sv)))
sv_max[samples] = list(
map(max, zip(sv_max[samples], sv)))
else:
#print(sv)
#print(sv_min.keys())
sv_min[samples] = sv
sv_max[samples] = sv
stats = make_block_stats(blocks)
if args.total:
output += " {:>7}".format(int(stats["total"]))
else:
output += "\n" + stats["str"]
print(output)
stats_total = make_block_stats(blocks_total)
if len(args.files) > 1:
output = "{:>25}:".format("TOTAL FOR {} FILES".format(len(args.files)))
if args.total:
output += " {:>7}".format(int(stats_total["total"]))
else:
output += "\n" + stats_total["str"]
print(output)
if args.sv:
print("Min and max values for SV vectors:")
for s in sorted(sv_min.keys()):
min_out = "SV_MIN ({} samples): ".format(s)
max_out = "SV_MAX ({} samples): ".format(s)
#print(sv_min)
#print(sv_max)
for mins in sv_min[s]:
min_out += "{:>2} ".format(mins)
for maxs in sv_max[s]:
max_out += "{:>2} ".format(maxs)
print(min_out)
print(max_out)