def crawler(filename: str, geno_file: str):
# goes through ambs/unam json file to ensure that the validity of ambs/unam groups
a_u_dict = fm.unjson_it(filename)
seq = fm.reads(geno_file)
max = len(seq)
u_offset = 0
a_offset = 0 # important to note that ambs measures length and not position
for una in a_u_dict:
# key = unam
# value = ambs
u_offset += int(una)
a_offset += (int(a_u_dict[una]) + int(una) - 1)
if u_offset != 0:
assert seq[u_offset] in ['A', 'C', 'G', 'T']
if u_offset < max:
assert seq[u_offset + 1] == 'N'
assert seq[a_offset] == 'N'
if a_offset < max:
assert seq[a_offset + 1] in ['A', 'C', 'G', 'T']
u_offset += int(a_u_dict[una])
return
def read_json_file(filename: str):
a_u_dict = fm.unjson_it(filename)
with open('ambs', 'w') as ambs:
with open('unam', 'w') as unam:
for key in a_u_dict:
unam.write(str(key) + '\n')
ambs.write(str(a_u_dict[key]) + '\n')
def read_trues_chr(true_file: str, chrs_file: str):
print("Reading data: ")
print("Reading True Address file: ", end="")
trues = read_byte_to_queue(true_file)
print("done.\nReading chrs file: ", end='')
chrs_dict = unjson_it(chrs_file)
print("done.")
return trues, chrs_dict
def read_dict(filename: str, filetype='pickle'):
if filetype.lower() == 'pickle':
return unpickle_dict(filename)
elif filetype.lower() == 'json':
return unjson_it(filename)
elif filetype.lower() == 'msgpack':
return msgunpack_dict(filename=filename)
else:
raise Exception('unknown file type')
def read_data(true_file: str, tops_file: str, lcps_file: str, chrs_file: str):
print("Reading data: ")
print("Reading True Address file: ", end="")
trues = read_byte_to_queue(true_file)
print("done.\nReading Tops file: ", end="")
tops = read_byte_to_queue(tops_file)
print("done.\nReading Unique Start file: ", end='')
lcps = read_byte_to_queue(lcps_file)
print("done.\nReading chrs file: ", end='')
chrs_dict = unjson_it(chrs_file)
print("done.")
return trues, tops, lcps, chrs_dict
def test_validity():
# assumes that mu's have already been json'ed
mu = fm.unjson_it("../src/c22_mu")
geno = fm.read_unambiguous("../data/22.fa")
myd = {}
for key in tqdm(mu, desc="checking uniqueness"):
seq = geno[int(key):mu[key] + 1]
assert seq not in myd
myd[seq] = 1
if key == "700":
myd["ACGT"]
import sys
import os
from tqdm import tqdm
sys.path.append(os.getcwd().split('uniquekmer')[0] + 'uniquekmer/src')
import file_manager as fm
filename = '../22_json_default_dict'
print('reading jsoned dict')
d = fm.unjson_it(filename)
print('read!')
print('reading genome')
genome = fm.reads('../data/22.fa')
print('read!')
ambs=0
tops=0
# checking how many tops are less than 100
for sa in tqdm(d, desc='checking dict'):
top = int(d[sa][1])
sa = int(sa)
string = genome[sa:sa+top]
if 'N' in string:
ambs+=1
if top < 100:
def temp():
c22 = unjson_it(filename="c22_mu")
l0 = list(c22.keys())
l1 = list(c22.values())
just_dump(l0, l1, fn="c22_mu_just_dump")
def _part_1(genome, past, s_array, args=None, print=print):
try:
# check if args.LCPfile exists
# if it does, read the pickle file instead of calculating new lcp
inv_suff = []
# ___________________________________________
print('\n_____________________________________')
print('PART 1: COMPUTE LCP ARRAY')
print('_____________________________________\n')
# ____________________________________________
# if user has specified a LCP pickle file that already exists
if args.lcpfile and os.path.isfile(path=args.lcpfile):
print("uniques file exists: ")
#print(args.lcpfile, '\n')
# TODO: change this as necessary
# hopefully start_uniques will be pickled/jsom/msgpacked in the future
#lcp = unpickle_dict(filename=args.lcpfile)
lcp = unjson_it(args.lcpfile)
# find out what format the lcp was pickled
if type(lcp) == dict or type(lcp) == OrderedDict:
key = lcp.keys()[0]
value = lcp[key]
if key > value and (100 >= value >= 20):
print("old lcp pickle was in format sa:lcp")
lcp = deque(lcp.values())
elif key < value and (100 >= key >= 20):
print("old lcp pickle was in format lcp:sa")
lcp = deque(lcp.keys())
else:
print("not sure what's going on here for sa_lcp dict")
raise KeyboardInterrupt
s_array = deque(s_array)
elif type(lcp) == list:
print('LCP file read as list')
print("uniques unpacked\n")
past = get_time(past, print=print)
print("Computing Unique Start Lengths")
# combine sa and lcp to form a dict with keys: sa, values: unique_starts
# TODO: creating OrderedDict consumes too much memory
filename = append_file_name('json_lcp')
if args.outfile:
filename = args.outfile
if args.inverse:
inv_suff = unjson_it(args.inverse)
else:
inv_suff = inverse1(s_array=s_array)
else:
if args.inverse:
inv_suff = unjson_it(args.inverse)
inv_suff, lcp = kasai(genome=genome,
inv_suff=inv_suff,
s_array=s_array,
print=print)
past = time.time()
# convert suffix array (list) to suffix array (deque) for increased efficiency
print('Completed.')
# json it
filename = append_file_name('json_lcp')
if args.outfile:
filename = append_file_name(args.outfile + 'json_lcp')
print('json\'ing lcp array to %s', filename)
json_it(data=lcp, filename=filename)
print('LCP json\'ed!')
past = get_time(past, print=print)
return past, lcp, inv_suff
except Exception as e:
raise
