def EGAAC(fastas, window=5, **kw):
if check_sequences.check_fasta_with_equal_length == False:
print(
'Error: for "EGAAC" encoding, the input fasta sequences should be with equal length. \n\n'
)
return 0
if window < 1:
print('Error: the sliding window should be greater than zero' + '\n\n')
return 0
if check_sequences.get_min_sequence_length(fastas) < window:
print(
'Error: all the sequence length should be larger than the sliding window :'
+ str(window) + '\n\n')
return 0
group = {
'alphaticr': 'GAVLMI',
'aromatic': 'FYW',
'postivecharger': 'KRH',
'negativecharger': 'DE',
'uncharger': 'STCPNQ'
}
groupKey = group.keys()
encodings = []
header = ['#', 'label']
for w in range(1, len(fastas[0][1]) - window + 2):
for g in groupKey:
header.append('SW.' + str(w) + '.' + g)
encodings.append(header)
for i in fastas:
name, sequence, label = i[0], i[1], i[2]
code = [name, label]
for j in range(len(sequence)):
if j + window <= len(sequence):
count = Counter(sequence[j:j + window])
myDict = {}
for key in groupKey:
for aa in group[key]:
myDict[key] = myDict.get(key, 0) + count[aa]
for key in groupKey:
code.append(myDict[key] / window)
encodings.append(code)
return encodings
def CKSNAP(fastas, gap=2, **kw):
if gap < 0:
print('Error: the gap should be equal or greater than zero' + '\n\n')
return 0
if check_sequences.get_min_sequence_length(fastas) < gap + 2:
print(
'Error: all the sequence length should be larger than the (gap value) + 2 = '
+ str(gap + 2) + '\n\n')
return 0
AA = kw['order'] if kw['order'] != None else 'ACGT'
encodings = []
aaPairs = []
for aa1 in AA:
for aa2 in AA:
aaPairs.append(aa1 + aa2)
header = ['#', 'label']
for g in range(gap + 1):
for aa in aaPairs:
header.append(aa + '.gap' + str(g))
encodings.append(header)
for i in fastas:
name, sequence, label = i[0], i[1], i[2]
code = [name, label]
for g in range(gap + 1):
myDict = {}
for pair in aaPairs:
myDict[pair] = 0
sum = 0
for index1 in range(len(sequence)):
index2 = index1 + g + 1
if index1 < len(sequence) and index2 < len(
sequence
) and sequence[index1] in AA and sequence[index2] in AA:
myDict[sequence[index1] +
sequence[index2]] = myDict[sequence[index1] +
sequence[index2]] + 1
sum = sum + 1
for pair in aaPairs:
code.append(myDict[pair] / sum)
encodings.append(code)
return encodings
def ENAC(fastas, window=5, **kw):
if check_sequences.check_fasta_with_equal_length == False:
print(
'Error: for "ENAC" encoding, the input fasta sequences should be with equal length. \n\n'
)
return 0
if window < 1:
print('Error: the sliding window should be greater than zero' + '\n\n')
return 0
if check_sequences.get_min_sequence_length(fastas) < window:
print(
'Error: all the sequence length should be larger than the sliding window :'
+ str(window) + '\n\n')
return 0
AA = kw['order'] if kw['order'] != None else 'ACGU'
encodings = []
header = ['#', 'label']
for w in range(1, len(fastas[0][1]) - window + 2):
for aa in AA:
header.append('SW.' + str(w) + '.' + aa)
encodings.append(header)
for i in fastas:
name, sequence, label = i[0], i[1], i[2]
code = [name, label]
for j in range(len(sequence)):
if j < len(sequence) and j + window <= len(sequence):
count = Counter(sequence[j:j + window])
for key in count:
count[key] = count[key] / len(sequence[j:j + window])
for aa in AA:
code.append(count[aa])
encodings.append(code)
return encodings
def CKSAAGP(fastas, gap=5, **kw):
if gap < 0:
print('Error: the gap should be equal or greater than zero' + '\n\n')
return 0
if check_sequences.get_min_sequence_length(fastas) < gap + 2:
print(
'Error: all the sequence length should be larger than the (gap value) + 2 = '
+ str(gap + 2) + '\n\n')
return 0
group = {
'alphaticr': 'GAVLMI',
'aromatic': 'FYW',
'postivecharger': 'KRH',
'negativecharger': 'DE',
'uncharger': 'STCPNQ'
}
AA = 'ARNDCQEGHILKMFPSTWYV'
groupKey = group.keys()
index = {}
for key in groupKey:
for aa in group[key]:
index[aa] = key
gPairIndex = []
for key1 in groupKey:
for key2 in groupKey:
gPairIndex.append(key1 + '.' + key2)
encodings = []
header = ['#', 'label']
for g in range(gap + 1):
for p in gPairIndex:
header.append(p + '.gap' + str(g))
encodings.append(header)
for i in fastas:
name, sequence, label = i[0], re.sub('-', '', i[1]), i[2]
code = [name, label]
for g in range(gap + 1):
gPair = generateGroupPairs(groupKey)
sum = 0
for p1 in range(len(sequence)):
p2 = p1 + g + 1
if p2 < len(sequence
) and sequence[p1] in AA and sequence[p2] in AA:
gPair[index[sequence[p1]] + '.' +
index[sequence[p2]]] = gPair[index[sequence[p1]] +
'.' +
index[sequence[p2]]] + 1
sum = sum + 1
if sum == 0:
for gp in gPairIndex:
code.append(0)
else:
for gp in gPairIndex:
code.append(gPair[gp] / sum)
encodings.append(code)
return encodings