/
main.py
232 lines (193 loc) · 11.7 KB
/
main.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
from classifier import *
from tr_parser import *
from metrics import *
import multiprocessing
import pdb
import operator
def classification_cycle(kernel,
ngram,
class_good_oases,
class_good_trinity,
reads_names,
reads_seq,
fileout_suffix):
#print ("thread started")
(vectorizer_o, classifier_o) = train_classifier(class_good_oases, reads_seq, kernel, ngram, ngram)
(vectorizer_t, classifier_t) = train_classifier(class_good_trinity, reads_seq, kernel, ngram, ngram)
classify(vectorizer_obj=vectorizer_o,
classifier_obj=classifier_o,
window_size=10000,
filename_out="results/oases_classified-" + str(ngram) + "-" + kernel + fileout_suffix,
reads_names=reads_names,
reads_seq=reads_seq)
classify(vectorizer_obj=vectorizer_t,
classifier_obj=classifier_t,
window_size=10000,
filename_out="results/trinity_classified-" + str(ngram) + "-" + kernel + fileout_suffix,
reads_names=reads_names,
reads_seq=reads_seq)
def validation_cycle(kernel,
ngram,
class_good_oases,
class_good_trinity,
reads_names,
reads_seq,
fileout_suffix):
#print ("thread started")
oases_training_set = class_good_oases[0:int(len(class_good_oases)*0.75)]
trinity_training_set = class_good_trinity[0:int(len(class_good_trinity)*0.75)]
oases_test_set = class_good_oases[int(len(class_good_oases)*0.75)+1:int(len(class_good_oases)-1)]
trinity_test_set = class_good_trinity[int(len(class_good_trinity)*0.75)+1:int(len(class_good_trinity)-1)]
(vectorizer_o, classifier_o) = train_classifier(oases_training_set, reads_seq, kernel, 2, ngram, kernel)
(vectorizer_t, classifier_t) = train_classifier(trinity_training_set, reads_seq, kernel, 2, ngram, kernel)
process_list = []
p = multiprocessing.Process(target=classify, args=(vectorizer_o,
classifier_o,
10000,
"validation/oases_classified-" + str(ngram) + "-" + kernel.keys()[0] + "-" + str(kernel[kernel.keys()[0]]) + "-" + fileout_suffix + "-must-fit",
reads_names,
oases_test_set))
process_list.append(p)
p.start()
p = multiprocessing.Process(target=classify, args=(vectorizer_t,
classifier_t,
10000,
"validation/trinity_classified-" + str(ngram) + "-" + kernel.keys()[0] + "-" + str(kernel[kernel.keys()[0]]) + "-" + fileout_suffix + "-must-fit",
reads_names,
trinity_test_set))
process_list.append(p)
p.start()
test_list = list(operator.xor(set(reads_seq), set(class_good_oases)))
p = multiprocessing.Process(target=classify, args=(vectorizer_o,
classifier_o,
10000,
"validation/oases_classified-" + str(ngram) + "-" + kernel.keys()[0] + "-" + str(kernel[kernel.keys()[0]]) + "-" + fileout_suffix + "-must-not-fit",
reads_names,
test_list[0:10000]))
process_list.append(p)
p.start()
test_list = list(operator.xor(set(reads_seq), set(class_good_trinity)))
p = multiprocessing.Process(target=classify, args=(vectorizer_t,
classifier_t,
10000,
"validation/trinity_classified-" + str(ngram) + "-" + kernel.keys()[0] + "-" + str(kernel[kernel.keys()[0]]) + "-" + fileout_suffix + "-must-not-fit",
reads_names,
test_list[0:10000]))
process_list.append(p)
p.start()
for p in process_list:
p.join()
def launch_classification_threads(class_good_oases,
class_good_trinity,
reads_names,
reads_seq,
purpose,
fileout_suffix,
is_threaded=True):
target = None
if purpose == "real-life":
target = classification_cycle
elif purpose == "validate":
print ("cross-validating")
fileout_suffix += "-validating"
target = validation_cycle
# for ngram in [2, 3, 4, 5, 6, 7, 8]:
kernels = [{'svc': 'rbf'}, {'svc': 'poly'}, {'svc': 'sigmoid'}, {'rfc': 2}, {'rfc': 4}, {'rfc': 10}, {'rfc': 100}, {'rfc': 200}, {'rfc': 1000}]
if is_threaded:
for kernel in kernels:
for ngram in [2, 3, 4, 5, 6, 7, 8]:
target(kernel,
ngram,
class_good_oases,
class_good_trinity,
reads_names,
reads_seq,
fileout_suffix)
def main():
compute_for_real = False
validate = True
for bottom_bound in [0.3, 0.4, 0.5, 0.6]:
for top_bound in [0.99, 0.95, 0.9]:
fileout_suffix = "-" + str(bottom_bound) + "-" + str(top_bound) + "-"
(reads_names, reads_seq) = get_reads("data/ag_1_GGCTAC_filtered.fastq")
if not reads_names or not reads_seq:
print ("reads have been read unsuccessfully")
(oases_alignment_data, trinity_alignment_data) = get_alignment_data("data/results_Oases.txt",
"data/results_Trinity.txt")
if not oases_alignment_data or not trinity_alignment_data:
print ("align have been read unsuccessfully")
(oases_reads_names, oases_transcripts_names, oases_reads_seq) = get_reads_for_assembler("data/results_oases.sam")
if not oases_reads_names or not oases_transcripts_names or not oases_reads_seq:
print ("reads for oases have been read unsuccessfully")
(trinity_reads_names, trinity_transcripts_names, trinity_reads_seq) = get_reads_for_assembler("data/results_trinity.sam")
if not trinity_reads_names or not trinity_transcripts_names or not trinity_reads_seq:
print ("reads for trinity have been read unsuccessfully")
(ref, oases_reads, oases_name_index, trinity_reads, trinity_name_index) = get_assemblies("data/ref_for_reads.fasta",
"data/Oases.fasta",
"data/Trinity.fasta")
if not ref or not oases_reads or not trinity_reads or not oases_name_index or not trinity_name_index:
print ("assemblies have been read unsuccessfully")
(oases_distances_pairs, trinity_distances_pairs) = get_distances("data/Similar_transkripts_Oases.txt",
"data/Similar_transkripts_Trinity.txt")
if oases_distances_pairs is None or trinity_distances_pairs is None:
print ("unsuccessful distance reading")
reference_reads_to_transcripts = make_index_reads_to_transcripts(reads_names,
reads_seq)
oases_reads_to_transcripts = make_index_reads_to_transcripts(oases_reads_names,
oases_transcripts_names)
trinity_reads_to_transcripts = make_index_reads_to_transcripts(trinity_reads_names,
trinity_transcripts_names)
oases_transcripts_to_reads = make_index_transcripts_to_reads(oases_name_index,
oases_transcripts_names,
oases_reads_names)
trinity_transcripts_to_reads = make_index_transcripts_to_reads(trinity_name_index,
trinity_transcripts_names,
trinity_reads_names)
oases_index_by_name = make_index_by_name(oases_name_index)
trinity_index_by_name = make_index_by_name(trinity_name_index)
#TODO: check if it's really class trinity
class_good_trinity = reads_for_class(oases_alignment_data,
oases_name_index,
oases_transcripts_to_reads,
trinity_reads_to_transcripts,
trinity_index_by_name,
trinity_distances_pairs,
trinity_alignment_data,
reference_reads_to_transcripts,
top_bound,
bottom_bound)
f = file("data/Class_Trinity.txt", "w")
pickle.dump(class_good_trinity, f)
f.close()
class_good_oases = reads_for_class(trinity_alignment_data,
trinity_name_index,
trinity_transcripts_to_reads,
oases_reads_to_transcripts,
oases_index_by_name,
oases_distances_pairs,
oases_alignment_data,
reference_reads_to_transcripts,
top_bound,
bottom_bound)
f = file("data/Class_Oases.txt", "w")
pickle.dump(class_good_oases, f)
f.close()
if compute_for_real:
launch_classification_threads(class_good_oases,
class_good_trinity,
reads_names,
reads_seq,
"real-life",
fileout_suffix)
if validate:
launch_classification_threads(class_good_oases,
class_good_trinity,
reads_names,
reads_seq,
"validate",
fileout_suffix)
start = time.time()
main()
print ('Completed in ' + str(int(time.time() - start)) + " seconds")
# import cProfile
#cProfile.run('main()')