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"]
def temp_forward_unique_check():
geno = read_unambiguous(filename=PATH + FORWARD)
# comment()
s_arr = read_byte_numpy(append_file_name('data/22.sa'))
inv_suff, lcp = kasai(geno, s_arr)
myd = {}
for num in range(len(s_arr)):
myd[s_arr[num]] = lcp[num]
#trues0 = list(get_uniques(lcp))
json_it(trues0, "c22_forward_uniques")
def SA_file(filename:str):
print('Reading file: ')
sequence = read_unambiguous(filename)
print('File read!\nCreating Suffix Array Naively:')
s_array, _ = naive_SA(sequence)
length = len(s_array)
s_array.insert(0,length)
print('Suffix Array Created!\nWriting to file: ')
write_array_to_byte(filename='fake_genome_sa',byte_arr=s_array)
return
def genome_reads_test(filename):
# testing reads() from file manager
filename = append_file_name(filename)
past = time.time()
read_bioseq = read_unambiguous(filename=filename)
current = time.time()
print('genome read with Bio.Seq. Time elapsed: ', current - past)
past = current
read_reads = reads(filename=filename)
current = time.time()
print('genome read with Reads.py. Time elapsed: ', current - past)
assert type(read_bioseq) is Bio.Seq.Seq
assert type(read_reads) is str
assert read_reads == str(read_bioseq)
def driver():
geno = read_unambiguous(TWO_WAY_GENO)
s_arr = read_byte_numpy(S_ARRAY)
def sesame_plant():
s_arr = read_byte_numpy(append_file_name('data/22.sa'))
geno = read_unambiguous(append_file_name('data/22.fa'))
se = SAGuide(s_arr=s_arr, geno=geno)
print(se)
def mu_driver():
"""
similar function as driver.py, except include minimal uniques instead of finding 20-100 uniquemers
:return:
"""
try:
# gitignore()
print('reading original genome: ', end='')
chrs, geno = chr_splits(filename=PATH + ORIGINAL)
json_it(chrs, append_file_name("json_chrs"))
del chrs
print('done.\nreading original SA...: ', end='')
s_arr = read_byte_numpy(append_file_name('data/genome.sa'))
lcp1 = kasai(geno, s_arr)[1]
d1 = OrderedDict(mu(SA=s_arr, LCP=lcp1))
del lcp1
del s_arr
au = _part_2(genome_file_name=PATH + ORIGINAL)
print("au list: ", list(au))
# *************************
# (2) flipped
# *************************
print("performing flips: ")
geno2 = read_unambiguous(PATH + FLIPPED)
s_arr2 = read_byte_numpy(append_file_name('data/flippedGeno.sa'))
lcp2 = kasai(geno2, s_arr2)[1]
del geno2
mu_result = dict(compare(d=d1, SA=s_arr2, LCP=lcp2))
del lcp2
mu_result = OrderedDict(sort_mu(mu_result))
mu_result = OrderedDict(true_address_dict(mu_result, au))
json_it(mu_result, append_file_name(files['MU_RESULT']))
#contigs = list(find_contigs(d=old_mu_result_without_true_addresses, bot=20, top=100))
contigs = OrderedDict(
find_perfect_contigs(d=mu_result, bot=20, top=100))
json_it(contigs, append_file_name(files['PERFECT_CONTIGS']))
contigs = list(within_distance(d=contigs, distance=300))
json_it(contigs,
append_file_name(files['PERFECT_CONTIGS_WITH_DISTANCE']))
print("number of contigs: ", len(contigs))
print("done")
except Exception as e:
raise
def efficient_mu_driver():
"""
NOTES:
07/05: You MUST run get_uniques first before sorting the lcp
:return:
"""
try:
# comment()
geno = reads(filename=PATH + FORWARD)
geno_length = len(geno)
# comment()
s_arr = read_byte_numpy(append_file_name('data/22.sa'))
inv_suff, lcp = kasai(geno, s_arr)
lcp = kasai(geno, s_arr)[1]
del geno, s_arr
# comment()
au = _part_2(genome_file_name=PATH + FORWARD)
lcp = list(get_uniques(lcp))
trues0 = list(sort_lcp(lcp=lcp, inv_suff=inv_suff))
del lcp
bad_address = forbiddens(inv_suff=inv_suff, lcp=trues0, au=au)
del inv_suff
geno = read_unambiguous(filename=PATH + FLIPPED)
s_arr = read_byte_numpy(append_file_name('data/f22.sa'))
inv_2, lcp = kasai(geno, s_arr)
lcp = kasai(geno, s_arr)[1]
del geno, s_arr
lcp = list(get_uniques(lcp))
trues1 = list(sort_lcp(lcp=lcp, inv_suff=inv_2))
del lcp, inv_2
# mu_s, mu_e = list(compare(inv0=inv_suff, trues0=trues0, inv1=inv_2, trues1=trues, bad_address=bad_address))
mu_s = []
mu_e = []
au_dict = {}
for item in list(au):
au_dict[item[0]] = item[1]
del au
u_ceil = list(au_dict)[0]
u_floor = 0
a_offset = au_dict[u_ceil]
# mu_s, mu_e = list(compare(trues0=trues0, trues1=trues1, bad_address=bad_address))
for tup in compare_no_inv_suff(trues0=trues0,
trues1=trues1,
bad_address=bad_address,
geno_length=geno_length):
sa = tup[0]
if sa < u_floor:
raise Exception(
"SA is less than u_floor. Possible that s_arr not sorted correctly?"
)
if sa > u_ceil and len(au_dict) > 1:
u_floor = u_ceil
del au_dict[u_ceil]
u_ceil = list(au_dict)[0]
a_offset = au_dict[u_ceil]
elif len(au_dict) < 1:
print("not au_dict reached")
break
# mu_s.append(tup[0])
mu_s.append(sa + a_offset)
mu_e.append(tup[1])
# TODO: 07/05 made the line below return a dict as well as accept geno
# to return to before, do not input geno and output two lists
# myd = dict(compare(trues0 = trues0, trues1 = trues1, bad_address=bad_address))
# json_it(data=myd, filename="c22_mu")
assert len(mu_s) == len(mu_e)
just_dump(myl=mu_s, fn="c22_mu_starts_0709", print_length=True)
just_dump(myl=mu_e, fn="c22_mu_ends_0709", print_length=True)
# 07/08: changed get_uniques so that it doesn't yield past or lcp + 1
# json_it(mu_s, "efficient_mu_starts")
# json_it(mu_e, "efficient_mu_ends")
# stitched = list(stitch(starts=mu_s, uniques=mu_e))
# json_it(stitched, "stitched")
# print("Number of stitched: " + str(len(stitched)))
# print("Number of MU: " + str(len(mu_s)))
# findmean(mys = mu_s, mye=mu_e)
except IndexError:
pass
except Exception:
print(traceback.format_exc())
breakpoint()