-
-
Notifications
You must be signed in to change notification settings - Fork 1k
/
msplicer
executable file
·355 lines (290 loc) · 10.8 KB
/
msplicer
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
#!/usr/bin/env python
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# Written (W) 2007 Gunnar Raetsch
# Written (W) 2006-2008 Soeren Sonnenburg
# Copyright (C) 2006-2008 Fraunhofer Institute FIRST and Max-Planck-Society
#
try:
import os
import os.path
import sys
import pickle
import bz2
import numpy
import optparse
import genomic
import model
import seqdict
import shogun
d=shogun.DynProg()
if (d.version.get_version_revision() < 2997):
print
print "ERROR: SHOGUN VERSION 0.6.2 or later required"
print
sys.exit(1)
from content_sensors import content_sensors
from signal_detectors import signal_detectors
from plif import plif
except ImportError, e:
print e
print
print "ERROR IMPORTING MODULES, MAKE SURE YOU HAVE SHOGUN INSTALLED"
print
sys.exit(1)
msplicer_version='v0.3'
class msplicer:
def __init__(self):
self.model = None
self.plif = None
self.signal = None
self.content = None
self.model_name = None
def load_model(self, filename):
self.model_name = filename
sys.stderr.write('loading model file\n')
f=None
picklefile=filename+'.pickle'
if os.path.isfile(picklefile):
self.model=pickle.load(file(picklefile))
else:
if filename.endswith('.bz2'):
f=bz2.BZ2File(filename);
else:
f=file(filename);
self.model=model.parse_file(f)
f.close()
f=file(picklefile,'w')
pickle.dump(self.model, f)
f.close()
self.plif=plif(self.model)
self.signal=signal_detectors(self.model)
self.content=content_sensors(self.model)
def compute_seqmatrix(self, seq):
# start-state: 0
# exon-start-state: 1
# donor-state: 2
# acceptor-state: 3
# exon-end-state: 4
# stop-state: 5
start_idx = numpy.where(self.model.statedescr == 0)[0]
exon_start_idx = numpy.where(self.model.statedescr == 1)[0]
don_idx = numpy.where(self.model.statedescr == 2)[0]
acc_idx = numpy.where(self.model.statedescr == 3)[0]
exon_stop_idx = numpy.where(self.model.statedescr == 4)[0]
stop_idx = numpy.where(self.model.statedescr == 5)[0]
# start positions
positions=[(0,0,start_idx)]
positions.append((seq.start,0,exon_start_idx))
# end positions
positions.append((seq.end, 0, exon_stop_idx[0]))
if len(exon_stop_idx)>1:
idx = numpy.where(numpy.array(seq.preds['acceptor'].positions,numpy.int32)==seq.end)[0]
if len(idx)==1:
positions.append((seq.end, seq.preds['acceptor'].scores[idx], exon_stop_idx[1]))
positions.append((len(seq.seq)-1,0,stop_idx))
# donor posititions
for i in don_idx:
positions.extend(zip(seq.preds['donor'].positions,
seq.preds['donor'].scores,
len(seq.preds['donor'].positions)*[i]))
# acceptor positions
for i in acc_idx:
positions.extend(zip(seq.preds['acceptor'].positions,
list(seq.preds['acceptor'].scores),
len(seq.preds['acceptor'].positions)*[i]))
positions.sort(cmp=lambda x,y : int(x[0]-y[0]))
unique_positions= numpy.unique(numpy.array([ x[0] for x in positions ], numpy.int32))
seqmatrix= -numpy.infty * numpy.ones((len(self.model.statedescr),len(unique_positions)))
for i in xrange(len(positions)):
p = numpy.where(positions[i][0]==unique_positions)[0] ;
assert(len(p)==1)
p = p[0] ;
seqmatrix[positions[i][2],p]=positions[i][1]
if len(don_idx)>1: # orf case
for i in xrange(len(unique_positions)):
if seqmatrix[don_idx[0], i] > -1e20:
s1 = seq.seq[unique_positions[i]-1:unique_positions[i]+1]
s2 = seq.seq[unique_positions[i]-2:unique_positions[i]+1]
if s1 in ['TG']: seqmatrix[don_idx[1], i]=-numpy.infty
if s1 not in ['TG']: seqmatrix[don_idx[2], i]=-numpy.infty
if s2 in ['TAG', 'TGG']: seqmatrix[don_idx[3], i]=-numpy.infty
if s2 not in ['TAG']: seqmatrix[don_idx[4], i]=-numpy.infty
if s2 not in ['TGG']: seqmatrix[don_idx[5], i]=-numpy.infty
if len(acc_idx)>1: # orf case
for i in xrange(len(unique_positions)):
if seqmatrix[acc_idx[0], i] > -1e20:
s1 = seq.seq[unique_positions[i]-1:unique_positions[i]+1]
s2 = seq.seq[unique_positions[i]-1:unique_positions[i]+2]
if s2 in ['GAA', 'GAG', 'GGA']: seqmatrix[acc_idx[2], i]=-numpy.infty
if s1 in ['GA', 'GG']: seqmatrix[acc_idx[4], i]=-numpy.infty
if s1 in ['GA']: seqmatrix[acc_idx[5], i]=-numpy.infty
plifstatemat = -numpy.ones((len(self.model.statedescr),1), numpy.int32);
plifstatemat[acc_idx,0] = 0 ; # acceptors use first plif
plifstatemat[don_idx,0] = 1 ; # donors use second plif
return (seqmatrix, unique_positions, plifstatemat)
def initialize_dynprog(self, seq):
dyn=shogun.DynProg()
self.content.initialize_content(dyn)
n=len(self.model.p)
dyn.set_num_states(n)
dyn.set_p_vector(self.model.p)
dyn.set_q_vector(self.model.q)
dyn.set_a_trans_matrix(self.model.a_trans)
#design scoring seqmatrix
(seqmatrix, positions, plifstatemat) = self.compute_seqmatrix(seq)
dyn.best_path_set_seq(seqmatrix)
dyn.best_path_set_pos(positions)
dyn.best_path_set_orf_info(self.model.orf_info)
dyn.best_path_set_plif_list(self.plif.get_plif_array())
dyn.best_path_set_plif_id_matrix(self.model.plifidmat.T)
dyn.best_path_set_plif_state_signal_matrix(plifstatemat)
s=[]; s+=seq.seq;
dyn.best_path_set_single_genestr(numpy.array(s))
dyn.best_path_set_dict_weights(self.content.get_dict_weights())
# self.precompute_content_svm_values(self, dyn, seq, positions)
return (dyn,positions)
#def precompute_content_svm_values(self, dyn, seq, positions):
# wordstr=dyn.create_word_string(seq, 1, len(seq));
# dyn.init_content_svm_value_array(Npos)
# weights = self.content.get_dict_weights()
# #n = size(weights, 1)
# #m = size(weights, 2)
# dyn.precompute_content_values(wordstr, positions, len(positions), len(seq), self.content.get_dict_weights(), n*m);
# dyn.set_genestr_len(len(seq));
# return (dyn)
def write_gff(self, outfile, pred, name, score, skipheader):
descr=list()
for i in xrange(pred.shape[0]):
d=dict()
d['seqname']=name
d['source']='msplicer'
d['feature']='exon'
d['start']=pred[i,0]+1
d['end']=pred[i,1]
d['score']=score
d['strand']='+'
d['frame']=0
descr.append(d)
genomic.write_gff(outfile, ('msplicer',msplicer_version + ' ' + self.model_name),
('DNA', name), descr, skipheader)
def predict_file(self, fname, (start,end)):
skipheader=False
fasta_dict = genomic.read_fasta(file(fname))
sys.stderr.write('found fasta file with ' + `len(fasta_dict)` + ' sequence(s)\n')
seqs= seqdict.seqdict(fasta_dict, (start,end))
#get donor/acceptor signal predictions for all sequences
self.signal.predict_acceptor_sites_from_seqdict(seqs)
self.signal.predict_donor_sites_from_seqdict(seqs)
for seq in seqs:
#initialize dynamic programming, with content sensors
#signal detectors, Plifs and HMM like model
(dyn,positions)=self.initialize_dynprog(seq)
#compute max likely path
dyn.best_path_call(1, self.model.use_orf)
scores=dyn.best_path_get_scores()
states=dyn.best_path_get_states()
pos=dyn.best_path_get_positions()
pred_states=states[0][0:numpy.where(pos[0]==-1)[0]][1:-1]
pred=positions[pos[0][0:numpy.where(pos[0]==-1)[0]][1:-1]]
#print scores
#print pred_states
#print pred
#print len(pred_states)
if (len(pred_states)>0):
if (pred_states[-1]==15): # joint state for acceptor and stop codon
pred_ = numpy.zeros(len(pred)+1, numpy.int32) ;
pred_[0:len(pred)] = pred ;
pred_[-1] = pred[-1]
pred = pred_
pred=pred.reshape((len(pred)/2,2))
self.write_gff(outfile, pred, seq.name, scores, skipheader)
skipheader=True
if 0:
my_posi = numpy.array([ 1, 400, 408, 451, 1188, 1785, 1858, 2732, 2924, 3869, 3948, 4348 ], numpy.int32)-1 ;
my_pos = numpy.zeros(len(my_posi), numpy.int32) ;
print positions, my_posi
for i in xrange(len(my_posi)):
my_pos[i] = numpy.where(positions == my_posi[i])[0]
my_states = numpy.array([0, 13, 6, 12, 2, 8, 4, 10, 4, 10, 14, 16], numpy.int32)
#my_pos = numpy.array([ 0, 51, 169, 204, 216, 241, 300, 355, 360, 397], numpy.int32) ;
#my_states = numpy.array([0, 3, 1, 2, 1, 2, 1, 2, 4, 5], numpy.int32)
my_states = states[0][0:numpy.where(pos[0]==-1)[0]]
my_pos = pos[0][0:numpy.where(pos[0]==-1)[0]]
print my_states
print my_pos
print positions[my_pos]
dyn.best_path_set_my_state_seq(my_states)
dyn.best_path_set_my_pos_seq(my_pos)
dyn.io.set_loglevel(shogun.M_DEBUG)
dyn.best_path_deriv_call()
def print_version():
sys.stderr.write('mSplicer '+msplicer_version+'\n')
def parse_options():
parser = optparse.OptionParser(usage="usage: %prog [options] seq.fa")
parser.add_option("-o", "--outfile", type="str", default='stdout',
help="File to write the results to")
parser.add_option("-v", "--version", default=False,
help="Show some more information")
parser.add_option("--start", type="int", default=499,
help="coding start (zero based, relative to sequence start)")
parser.add_option("--stop", type="int", default=-499,
help="""coding stop (zero based, if positive relative to
sequence start, if negative relative to sequence end)""")
parser.add_option("--model", type="str", default='WS160',
help="mSplicer Model to use in predicting")
(options, args) = parser.parse_args()
if options.version:
print_version()
sys.exit(0)
if len(args) != 1:
parser.error("incorrect number of arguments")
fafname=args[0]
if not os.path.isfile(fafname):
parser.error("fasta file does not exist")
if options.model.endswith('gc'):
gc=1
model=options.model[:-2]
else:
gc=0
model=options.model
if model.startswith('orf'):
orf=1
model=model[3:]
else:
orf=0
modelfname = 'data/msplicer_elegans%s_gc=%d_orf=%d.dat.bz2' % (model, gc, orf)
print "loading model file " + modelfname,
if not os.path.isfile(modelfname):
print "...not found!\n"
parser.error("""model should be one of:
WS120, WS120gc, orfWS120, WS150,
WS160, WS160gc, orfWS160gc
""")
if options.outfile == 'stdout':
outfile=sys.stdout
else:
try:
outfile=file(options.outfile,'w')
except IOError:
parser.error("could not open %s for writing" % options.outfile)
if options.start<80:
parser.error("--start value must be >=80")
if options.stop > 0 and options.start >= options.stop - 80:
parser.error("--stop value must be > start + 80")
if options.stop < 0 and options.stop > -80:
parser.error("--stop value must be <= - 80")
# shift the start and stop a bit
options.start -= 1 ;
options.stop -= 1 ;
return ((options.start,options.stop), fafname, modelfname, outfile)
if __name__ == '__main__':
dyn=shogun.DynProg()
(startstop, fafname, modelfname, outfile ) = parse_options()
p=msplicer()
p.load_model(modelfname);
p.predict_file(fafname, startstop)