-
Notifications
You must be signed in to change notification settings - Fork 0
/
generate_features.py
390 lines (344 loc) · 12.3 KB
/
generate_features.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
import os
import numpy as np
import pandas as pd
from multiprocessing import Pool
from generate_pssm import *
# from Bio import SeqIO
aa_alphabet = ['A', 'R', 'N', 'D', 'C', 'Q', 'E', 'G', 'H', 'I', 'L', 'K', 'M', 'F', 'P', 'S', 'T', 'W', 'Y', 'V']
root_path = ''
class Seq():
def __init__(self, id, seq):
self.name = id[1:]
self.seq = seq
self.id =id[1:]
def parser_fasta(fasta):
ids, seqs = load_seqs(fasta)
res = []
for i in range(len(ids)):
res.append(Seq(ids[i], seqs[i]))
return res
def read_fasta(fn):
"""
read fasta file
:param fn: file name
:return: AA sequences, AA names
"""
seq_names = []
seqs = []
with open(fn, 'r') as f:
lines = f.readlines()
for line in lines:
line = line.strip()
if line[0] == '>':
# 去掉'|'后的标签,因为可能是程序自己加的
if '|' in line:
line = line.split('|')[0]
seq_names.append(line)
else:
seqs.append(line)
return seqs, seq_names
def gen_features(fn, pt, feature_name, pssm_file = None):
"""
:param fn: file name
:param pt: peptide type
:param feature_name: target feature name
:param pssm_file: pssm file path, some features need generate pssm matrix first
:return: generate feature files under user folder
"""
print('--------------generating feature', feature_name,'--------------')
if feature_name == 'AAC':
return AAC(fn, pt)
# elif feature_name == 'BIT20':
# return BIT20NT4(fn, pt)
elif feature_name == 'Kmer':
return Kmer(fn, pt)
elif feature_name == 'Top_n_gram':
return Top_n_gram(fn, pt)
elif feature_name == 'DP':
return DP(fn, pt)
elif feature_name == 'DT':
return DT(fn, pt)
elif feature_name == 'DR':
return DR(fn, pt)
elif feature_name == 'PseAAC':
return PseAAC(fn, pt)
else:
feature = Feature()
feature.get_dbt_feature(fn , feature_name, pt, pssm_file)
def AAC(fn, pt):
"""
AAC: Amino acid component
:param fn: user folder name
:param pt: peptide name
:return: list of AAC
"""
seqs, _ = read_fasta(fn + '/test.fasta')
res = []
aa_dict = {}
for i in range(len(aa_alphabet)):
aa_dict[aa_alphabet[i]] = 0
for seq in seqs:
tmp = aa_dict.copy()
n = len(seq)
for a in seq:
tmp[a] += 1
for a in aa_alphabet:
tmp[a] /= n
res.append(list(tmp.values()))
df = pd.DataFrame(res)
df.to_csv(fn + '/' + pt + '_AAC.txt', header=None, index=None, sep=',')
return res
def Top_n_gram(fn, pt):
"""
creating Top-n-gram matirx under user dir
"""
cmd = 'python ' + root_path + 'tool/BioSeq-Analysis2/profile.py ' + \
fn + '/test.fasta -method Top-n-gram -f csv -out ' + \
fn + '/' + pt + '_Top-n-gram.txt -n 2'
os.system(cmd)
def DP(fn, pt):
"""
creating DP matirx under user dir
"""
cmd = 'python ' + root_path + 'tool/BioSeq-Analysis2/nac.py ' + \
fn + '/test.fasta Protein DP -f csv -out ' + \
fn + '/' + pt + '_DP.txt'
os.system(cmd)
def DT(fn, pt):
"""
creating DT matirx under user dir
"""
cmd = 'python ' + root_path + 'tool/BioSeq-Analysis2/profile.py ' + \
fn + '/test.fasta -method DT -f csv -out ' + \
fn + '/' + pt + '_DT.txt -max_dis 3'
os.system(cmd)
def DR(fn, pt):
"""
creating DR matirx under user dir
"""
cmd = 'python ' + root_path + 'tool/BioSeq-Analysis2/nac.py ' + \
fn + '/test.fasta Protein DR -f csv -out ' + \
fn + '/' + pt + '_DR.txt'
os.system(cmd)
def Kmer(fn, pt):
"""
creating Kmer matirx under user dir
"""
cmd = 'python ' + root_path + 'tool/BioSeq-Analysis2/nac.py ' + \
fn + '/test.fasta Protein Kmer -f csv -out ' + \
fn + '/' + pt + '_Kmer.txt -k 2'
os.system(cmd)
def PseAAC(fn, pt):
"""
creating PseAAC matirx under user dir
"""
cmd = 'python ' + root_path + 'tool/BioSeq-Analysis2/pse.py ' + fn + '/test.fasta Protein -method PC-PseAAC-General ' + \
'-f csv -out ' + fn + '/' + pt + '_PseAAC.txt -lamada 2 -w 0.3 -labels -1'
os.system(cmd)
def gen_peps(fn, pt):
"""
creating Bit20NTCT, CTD, GAP5, Ngram(N=1)1, OVNT(5)84
"""
print('--------------generating feature', 'BIT CTD GAP5 Ngram OVNT(5)84', '--------------')
cmd = 'python ' + root_path + 'gen_pepred_features.py -src ' + fn + ' -pt ' + pt
os.system(cmd)
class Feature():
"""
Generate features including PPCT, PSSM_DT and PSFM_DBT and save them under user idr
"""
co = 0
def __init__(self):
co = 0
self.seq_len = 700
self.coding = ['A', 'R', 'N', 'D', 'C', 'Q', 'E', 'G', 'H', 'I', 'L',
'K', 'M', 'F', 'P', 'S', 'T', 'W', 'Y', 'V']
self.blosum62 = {}
# TODO
# blosum_reader = open('./tool/psiblast/blosum62', 'r')
blosum_reader = open(root_path + 'tool/psiblast/blosum62', 'r')
count = 0
for line in blosum_reader:
count = count + 1
if count <= 7:
continue
line = line.strip('\r').split()
self.blosum62[line[0]] = [float(x) for x in line[1:21]]
def get_protein_blosum(self, protein):
seq_str = str(protein.seq).replace('X', '')
seq_str = seq_str.replace('B', '')
seq_str = seq_str.replace('Z', '')
seq_str = seq_str.replace('U', '')
seq_str = seq_str.replace('J', '')
seq_str = seq_str.replace('O', '')
protein_lst = []
for aa in seq_str:
aa = aa.upper()
protein_lst.append(self.blosum62[aa])
return np.array(protein_lst)
def generate_protein_psfm(self, protein):
seq_str = str(protein.seq).replace('X', '')
seq_str = seq_str.replace('B', '')
seq_str = seq_str.replace('Z', '')
seq_str = seq_str.replace('U', '')
seq_str = seq_str.replace('J', '')
seq_str = seq_str.replace('O', '')
mat=np.zeros([len(seq_str),20])
i=0
for aa in seq_str:
aa = aa.upper()
indx = self.coding.index(aa)
mat[i][indx]=1.0
i += 1
return mat
def one_hot(self, file_path):
# seq_record = list(SeqIO.parse(file_path, 'fasta'))
seq_record = parser_fasta(file_path)
N=len(seq_record)
mats = np.asarray(
np.zeros([N, self.seq_len, 20]))
i=0
for prot in seq_record:
if len(prot.seq)<self.seq_len:
seqLen=len(prot.seq)
else:
seqLen=self.seq_len
for j in range(seqLen):
e=list(prot.seq)[j]
indx=self.coding.index(e)
mats[i][j][indx]=1.0
i+=1
return mats
def read_pssm(self, pssm_file):
with open(pssm_file, 'r') as f:
lines = f.readlines()
lines = lines[3:-6]
pro_seq=[]
mat = []
for line in lines:
tmp = line.strip('\n').split()
if len(tmp)==0:
break
pro_seq.append(tmp[1])
tmp = tmp[2:22]
mat.append(tmp)
mat = np.array(mat)
mat = mat.astype(float)
return pro_seq, mat
def read_psfm(self, pssm_file):
with open(pssm_file, 'r') as f:
lines = f.readlines()
lines = lines[3:-6]
pro_seq=[]
mat = []
for line in lines:
tmp = line.strip('\n').split()
if len(tmp)==0:
break
pro_seq.append(tmp[1])
tmp = tmp[22:42]
mat.append(tmp)
mat = np.array(mat)
mat = mat.astype(float)
mat = np.divide(mat, 100)
return pro_seq,mat
def average(self, matrixSum, seqLen):
matrix_array = np.array(matrixSum)
matrix_array = np.divide(matrix_array, seqLen)
matrix_array_shp = np.shape(matrix_array)
matrix_average = [(np.reshape(matrix_array, (matrix_array_shp[0] * matrix_array_shp[1],)))]
return matrix_average
def sigmoid(self, x):
s = 1 / (1 + np.exp(-x))
return s
def preHandleColumns(self, PSFM, PSSM, STEP, feature):
PSSM=np.asarray(PSSM, float)
PSFM=np.asarray(PSFM, float)
mat = np.zeros((20,20),float)
seq_cn = np.shape(PSSM)[0]
for i in range(20):
for j in range(20):
for k in range(seq_cn - STEP):
if feature == 'PSSM_DT' : mat[i][j] += (PSSM[k][i] * PSSM[k + STEP][j])
elif feature == 'PSFM_DBT': mat[i][j] += (PSFM[k][i] * PSFM[k + STEP][j])
elif feature == 'PPCT': mat[i][j] += (PSSM[k][i] * PSFM[k + STEP][j] + PSFM[k][i]* PSSM[k+STEP][j] +PSFM[k][i] * PSFM[k + STEP][j] +PSSM[k][i] * PSSM[k + STEP][j])
return mat
def psfm_dbt(self, PSFM, PSSM, end, feature):
seq_cn = float(np.shape(PSSM)[0])
vector = []
for i in range(0, end + 1):
matrix = self.preHandleColumns(PSFM, PSSM, i, feature)
ksb_vector = self.average( matrix, float(seq_cn - i))
vector += list(ksb_vector[0])
return vector
#PSSM-DT PSFM-DBT PPCT
def get_dbt_feature(self, source_path, feature, dataset, pssm_dir=None):
file_path = source_path + '/test.fasta'
print('Reading:', file_path)
save_path = source_path + '/' + dataset + '_' + feature + '.txt'
if feature == 'PSSM_DT': end = 5
elif feature == 'PSFM_DBT': end = 4
elif feature == 'PPCT': end = 4
print('save_path:', save_path)
read_count = 0
generate_count = 0
if os.path.isfile(save_path):
mats = np.loadtxt(save_path, dtype = np.float)
return mats
else:
# seq_record = list(SeqIO.parse(file_path, 'fasta'))
seq_record = parser_fasta(file_path)
mats = []
n = len(seq_record)
count = 0
for prot in seq_record:
n -= 1
if '|' in prot.name:
pssm_path = pssm_dir + '/' + (prot.name).split('|')[0]+'_'+(prot.name).split('|')[1] + '.pssm'
else:
pssm_path = pssm_dir + '/' + prot.name + '.pssm'
if os.path.isfile(pssm_path):
read_count += 1
pro_seq, pssm_profile=self.read_pssm(pssm_path)
pro_seq, psfm_profile = self.read_psfm(pssm_path)
print('read pssm '+ str(prot.name) + ': ' +str(n))
else:
generate_count += 1
print('generate pssm '+prot.name + ': ' +str(n))
pssm_profile = self.get_protein_blosum(prot)
psfm_profile = self.generate_protein_psfm(prot)
vec = self.psfm_dbt(psfm_profile, pssm_profile,end,feature)
mats.append(vec)
print('mats', np.array(mats).shape)
np.savetxt(save_path, mats, fmt ='%.12f')
print('read',read_count,'generate',generate_count)
if __name__ == '__main__':
fn = 'test/'
pt = 'AAP'
gen_peps(fn, pt)
#
# #test Kmer
# gen_features(fn, pt, 'Kmer')
#
# #test DR
# gen_features(fn, pt, 'DR')
#
# test PseAAC
# gen_features(fn, pt, 'PseAAC')
#
# # test Top-n-gram
# gen_features(fn, pt, 'Top_n_gram')
#
# # test DT
# gen_features(fn, pt, 'DT')
#
# # test DP
# gen_features(fn, pt, 'DP')
# gen_pssm(fn, fn + '/profile')
#
# pssm_path = '../tmp/test/profile/pssm'
# # test PPCT
# gen_features(fn, pt, 'PPCT', pssm_path)
# # test PSSM_DT
# gen_features(fn, pt, 'PSSM_DT', pssm_path)
# # test PSFM_DBT
# gen_features(fn, pt, 'PSFM_DBT', pssm_path)