def generate(X, seqType, args):
'''
# Note-1: args.gGap --> 1, 2, 3
# Note-2: gGap --> ('X', 'X')
:param X:
:param seqType:
:param args:
:return:
'''
elements = utils.sequenceElements(seqType)
m2 = list(itertools.product(elements, repeat=2))
m = m2
# print(args.gGap)
T = []
for x in X:
x = x[:args.terminusLength]
t = []
for i in range(1, args.gGap + 1, 1):
V = utils.kmers(x, i + 2)
# seqLength = len(x) - (i+2) + 1
for gGap in m:
# print(gGap[0], end='')
# print('-'*i, end='')
# print(gGap[1])
# trackingFeatures.append(gGap[0] + '-' * i + gGap[1])
C = 0
for v in V:
if v[0] == gGap[0] and v[-1] == gGap[1]:
C += 1
# print(C, end=',')
t.append(C)
#end-for
#end-for
t = np.array(t)
# t = t.reshape(-1, 1)
T.append(t)
# end-for
T = np.array(T)
# print(T.shape)
totalFeature = 0
if seqType == 'DNA' or seqType == 'RNA':
totalFeature = (4 * args.gGap * 4)
else:
if seqType == 'PROT':
totalFeature = (20 * args.gGap * 20)
else:
None
#end-if
save.datasetSave(T, totalFeature, 'fg11')
#end-def
def log_ZS_analytic((matrix, mu, Ne)):
"""compute log_Z analytically"""
acc = 0
nu = Ne - 1
L = len(matrix)
for kmer in kmers(L):
ep = score_seq(matrix, "".join(kmer))
acc += (1 / (1 + exp(ep - mu)))**(Ne - 1)
return log(acc)
def log_ZS_analytic((matrix, mu, Ne)):
"""compute log_Z analytically"""
acc = 0
nu = Ne - 1
L = len(matrix)
for kmer in kmers(L):
ep = score_seq(matrix, "".join(kmer))
acc += (1/(1+exp(ep-mu)))**(Ne-1)
return log(acc)
def log_Zb_chem_pot_ref_dep(L, sigma, G, mu, upto=4):
sites = kmers(L)
scores = [sigma * (L - site.count("A")) for site in sites]
Zs = [
sum([
exp(-sum(comb) - mu * k)
for comb in itertools.combinations(scores, k)
]) for k in trange(upto)
]
Z0 = sum(Zs)
return log(G / (4**L) * Z0)
def generate(X, seqType, args):
'''
:param X:
:param seqType:
:param args:
:return:
'''
if seqType == 'DNA' or seqType == 'RNA':
p = [0] * (4 * 4) # As we are working for g11
else:
if seqType == 'PROT':
p = [0] * (20 * 20) # As we are working for g11
else:
None
# Trail: Merged
elements = utils.sequenceElements(seqType)
m = list(itertools.product(elements, repeat=2))
T = []
for x in X:
merged = []
x = x[:args.terminusLength]
for i in range(1, args.gGap + 1):
kmers = utils.kmers(x, 2 + i) # g11 --> 2, gGap (g11+gGap)
t = []
require = (args.terminusLength - (2 + 1) + 1) - (len(x) -
(2 + i) + 1)
for kmer in kmers:
d = {''.join(_): 0 for _ in m}
segment = kmer[0] + kmer[-1]
d[segment] = 1
t.append(list(d.values()))
# break
# break
# print(v)
if require > 0:
for i in range(require):
t.append(p)
# end-for
else:
None
t = np.array(t)
# print(t)
merged.append(t)
# print('------------------')
# end-for
T.append(np.concatenate((merged), axis=1))
# end-for
T = np.array(T)
# print(T.shape)
totalFeature = 0
if seqType == 'DNA' or seqType == 'RNA':
totalFeature = (4 * args.gGap * 4)
else:
if seqType == 'PROT':
totalFeature = (20 * args.gGap * 20)
else:
None
# end-if
save.datasetSave(T, totalFeature, 'pg11')
#end-for
def compute_Z(model):
k = len(model)
L = int(1 + sqrt(1+8*k)/2)
return sum(exp(score(model, "".join(kmer))) for kmer in kmers(L))
ep = score_seq(matrix, "".join(kmer))
acc += (1 / (1 + exp(ep - mu)))**(Ne - 1)
return log(acc)
def log_ZM_analytic((matrix, mu, Ne), N):
log_ZS = log_ZS_analytic((matrix, mu, Ne))
return N * log_ZS
def log_Z_analytic((matrix, mu, Ne), N):
"""compute log_Z analytically"""
acc = 0
nu = Ne - 1
L = len(matrix)
for kmer in kmers(L):
ep = score_seq(matrix, "".join(kmer))
acc += (1 / (1 + exp(ep - mu)))**(Ne - 1)
return N * log(acc)
def log_ZS_naive((matrix, mu, Ne), trials=1000):
acc = 0
nu = Ne - 1
L = len(matrix)
for i in xrange(trials):
ep = score_seq(matrix, random_site(L))
acc += (1 / (1 + exp(ep - mu)))**(Ne - 1)
mean_Zs = acc / trials
return L * log(4) + log(mean_Zs)
gen_kmers <k> sample <sample name>
or
gen_kmers <k> genomes <opt:full>
"""
k = int(sys.argv[1])
version = sys.argv[2]
if version == 'sample':
sample_name = sys.argv[3]
filename = 'samples/%s.txt' % sample_name
sample_kmers = kmer_store()
with open(filename) as f:
for read in f:
read_kmers = kmers(read.strip(), k)
for kmer in read_kmers:
sample_kmers.update(kmer)
output_filename = 'pickles/%s_kmers_%d.pickle' % (os.path.basename(
os.path.normpath(filename)).replace('.txt', ''), k)
with open(output_filename, 'w') as f:
cPickle.dump(sample_kmers.kmers, f)
elif version == 'genomes':
full = True if (len(sys.argv) == 4 and sys.argv[3] == 'full') else False
kmer_spectra = defaultdict(lambda: [0] * 20)
for index, genome_filename in enumerate(
progress(
filter(lambda x: x.endswith('.fna'), os.listdir('genomes')))):
kmer_spectrum = {} if full else kmer_store()
def generate(X, seqType, args):
'''
:param X:
:param seqType:
:param args:
:return:
'''
if seqType == 'DNA' or seqType == 'RNA':
p = [0] * (4**args.kTuple)
else:
if seqType == 'PROT':
p = [0] * (20**args.kTuple)
else:
None
# print(p)
# print(len(p))
elements = utils.sequenceElements(seqType)
m = list(itertools.product(elements, repeat=args.kTuple))
terminusLength = args.terminusLength
# print(terminusLength)
T = []
for x in X:
# print(len(x))
x = x[:terminusLength]
# print(len(x))
# print('-----------------')
require = (terminusLength - args.kTuple + 1) - (len(x) - args.kTuple +
1)
# print(require)
t = []
kmers = utils.kmers(x, args.kTuple)
for kmer in kmers:
d = {''.join(i): 0 for i in m}
d[kmer] = 1
t.append(list(d.values()))
#end-for
if require > 0:
for i in range(require):
t.append(p)
#end-for
else:
None
t = np.array(t)
# print(t.shape)
T.append(t)
# print(t.shape)
#end-for
T = np.array(T)
# print(T.shape)
totalFeature = 0
if seqType == 'DNA' or seqType == 'RNA':
totalFeature = (4**args.kTuple)
else:
if seqType == 'PROT':
totalFeature = (20**args.kTuple)
else:
None
# end-if
save.datasetSave(T, totalFeature, 'pkmer')
#end-def
gen_kmers <k> sample <sample name>
or
gen_kmers <k> genomes <opt:full>
"""
k = int(sys.argv[1])
version = sys.argv[2]
if version == 'sample':
sample_name = sys.argv[3]
filename = 'samples/%s.txt' % sample_name
sample_kmers = kmer_store()
with open(filename) as f:
for read in f:
read_kmers = kmers(read.strip(), k)
for kmer in read_kmers:
sample_kmers.update(kmer)
output_filename = 'pickles/%s_kmers_%d.pickle' % (os.path.basename(os.path.normpath(filename)).replace('.txt',''), k)
with open(output_filename, 'w') as f:
cPickle.dump(sample_kmers.kmers, f)
elif version =='genomes':
full = True if (len(sys.argv) == 4 and sys.argv[3] == 'full') else False
kmer_spectra = defaultdict(lambda:[0]*20)
for index, genome_filename in enumerate(progress(filter(lambda x: x.endswith('.fna'), os.listdir('genomes')))):
kmer_spectrum = {} if full else kmer_store()
for kmer in kmers(nucleotides_fna('genomes/'+genome_filename), k):
if full:
kmer_spectrum[kmer] = kmer_spectrum[kmer]+1 if kmer in kmer_spectrum else 1
L = len(matrix)
for kmer in kmers(L):
ep = score_seq(matrix, "".join(kmer))
acc += (1/(1+exp(ep-mu)))**(Ne-1)
return log(acc)
def log_ZM_analytic((matrix, mu, Ne), N):
log_ZS = log_ZS_analytic((matrix, mu, Ne))
return N * log_ZS
def log_Z_analytic((matrix, mu, Ne), N):
"""compute log_Z analytically"""
acc = 0
nu = Ne - 1
L = len(matrix)
for kmer in kmers(L):
ep = score_seq(matrix, "".join(kmer))
acc += (1/(1+exp(ep-mu)))**(Ne-1)
return N * log(acc)
def log_ZS_naive((matrix, mu, Ne), trials=1000):
acc = 0
nu = Ne - 1
L = len(matrix)
for i in xrange(trials):
ep = score_seq(matrix, random_site(L))
acc += (1/(1+exp(ep-mu)))**(Ne-1)
mean_Zs = acc / trials
return L * log(4) + log(mean_Zs)
def log_ZM_naive((matrix, mu, Ne), N, trials=1000):
def compute_Z(model):
k = len(model)
L = int(1 + sqrt(1 + 8 * k) / 2)
return sum(exp(score(model, "".join(kmer))) for kmer in kmers(L))