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 generate(X, seqType, args):
'''
# Reference-1: (http://rosalind.info/glossary/k-mer-composition/) # It is also called "k-mer composition".
# Reference-2: iRecSpot-EF: https://www.sciencedirect.com/science/article/abs/pii/S0010482518302981
:param X:
:param seqType:
:param args:
:return:
'''
elements = utils.sequenceElements(seqType)
# print(elements)
# print(args.gGap)
# print(args.kTuple)
T = []
for x in X:
x = x[:args.terminusLength]
t = []
for i in range(1, args.kTuple + 1, 1):
v = list(itertools.product(elements, repeat=i))
# seqLength = len(x) - i + 1
for i in v:
# print(x.count(''.join(i)), end=',')
t.append(x.count(''.join(i)))
### --- ###
t = np.array(t)
# t = t.reshape(-1, 1)
# print(t.shape)
T.append(t)
#end-for
T = np.array(T)
# print(T.shape)
totalFeature = 0
if seqType == 'DNA' or seqType == 'RNA':
totalFeature = np.sum([4**(i) for i in range(1, args.kTuple + 1)])
else:
if seqType == 'PROT':
totalFeature = np.sum([20**(i) for i in range(1, args.kTuple + 1)])
else:
None
#end-if
save.datasetSave(T, totalFeature, 'fkmer')
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 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