-
Notifications
You must be signed in to change notification settings - Fork 0
/
simulate.py
555 lines (445 loc) · 14 KB
/
simulate.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
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
import os
import pprint
import random
random.seed(12345)
CHRLEN = {"chr2": 242193529,
"chr8": 145138636,
"chr12": 133275309}
# allowed chromosomal regions
CHRRANGES = {"chr2": (4880899, 83143574),
"chr8": (50000000, 106259331),
"chr12": (48581298, 120529644)}
# n = number of fragments
# single_chr = intrachromosomal (True), interchromosomal (False)
N = "n"
NEIGHBOR = "neighbor"
MIN_TANDEM_DUPLICATION = "min_tandem_duplication"
SINGLE_CHR = "single_chr"
WEIGHT_CHR = "weights_chr"
P_DEL_LEFT = "p_del_left"
P_DEL_RIGHT = "p_del_right"
P_INVERT = "p_invert"
P_DUP = "p_dup"
P_FOLDBACK = "p_foldback"
P_RETURN = "p_return"
P_CHOOSE_RANDOM = "p_choose_random"
# variables included in the metainformation string
N_FRAG = "fragments"
FRAG_LEN = "frag_len"
READ_LEN = "read_len"
SMALL_DEL = "small_del"
DUP = "dup"
INV = "inv"
INTERCHR = "interchromosomal"
MULTI_REGION = "multi-region"
FOLDBACK = "foldback"
RETURN = "return" # new fragment is next to the other
N_FRAG_POS = 0
FRAG_LEN_POS = 1
# READ_LEN_POS = 1
SMALL_DEL_POS = 2
DUP_POS = 3
INV_POS = 4
INTERCHR_POS = 5
MULTI_REGION_POS = 6
FOLDBACK_POS = 7
RETURN_POS = 8
CHR = 0
START = 1
END = 2
STRAND = 3
EVENTS = {
N_FRAG: N_FRAG_POS,
FRAG_LEN: FRAG_LEN_POS,
SMALL_DEL: SMALL_DEL_POS,
DUP: DUP_POS,
INV: INV_POS,
INTERCHR: INTERCHR_POS,
MULTI_REGION: MULTI_REGION_POS,
FOLDBACK: FOLDBACK_POS,
RETURN: RETURN_POS
}
CONF_INIT = {
N: 3,
NEIGHBOR: False,
SINGLE_CHR: True,
MIN_TANDEM_DUPLICATION: 1000,
WEIGHT_CHR: [50, 50],
P_DEL_LEFT: 20,
P_DEL_RIGHT: 30,
P_INVERT: 40,
P_DUP: 10,
P_FOLDBACK: 5,
P_RETURN: 5,
# probl to choose random position or otherwise transit to foldback or return
P_CHOOSE_RANDOM: 30
}
pp = pprint.PrettyPrinter(indent=4)
def generate_binary(length, l, all):
if length > 0:
generate_binary(length - 1, l + [0], all)
generate_binary(length - 1, l + [1], all)
else:
all.append(l)
def generate_conformation():
"""
Generate all possible conformation which we want to simulate
"""
# N_FRAG_POS = 0
# FRAG_LEN_POS = 1
# SMALL_DEL_POS = 2
# DUP_POS = 3
# INV_POS = 4
# INTERCHR_POS = 5
# MULTI_REGION_POS = 6
# FOLDBACK_POS = 7
# RETURN_POS = 8
# initialize simple excision
all_conf = [[1, 0, 0, 0, 0, 0, 0, 0, 0]]
values = {N_FRAG: list(range(1, 10)),
SMALL_DEL: [0, 1],
DUP: [0, 1],
INV: [0, 1],
INTERCHR: [0, 1],
MULTI_REGION: [0, 1],
FOLDBACK: [0, 1],
RETURN: [0, 1]
}
bins = []
generate_binary(6, [], bins)
for f in range(2, 11):
for b in bins:
# [#frag, len] + [small_del, dup, inv, interch, multi+region, foldback] + [return]
newconf = [f,0]+b+[0]
all_conf.append(newconf)
return all_conf
def code_metainformation(topology):
"""
Configuration string for every ecDNA
Convert configuration dictionary to a meta information array.
Args:
topology (dict): Contains all the configuration values
"""
# array of information
# pos 0 - read_len {0,1,2,3} - 0 - << 7000bp, 1 - ~7000, 2 - >> 7000, 3
#
arr = [0] * 9
arr[DUP_POS] = 1 if topology[DUP] > 0 else 0
arr[INV_POS] = 1 if topology[INV] > 0 else 0
arr[INTERCHR_POS] = 1 if topology[INTERCHR] > 0 else 0
arr[MULTI_REGION_POS] = 1 if topology[MULTI_REGION] > 0 else 0
arr[FOLDBACK_POS] = 1 if topology[FOLDBACK] > 0 else 0
return "".join([str(v) for v in arr])
def is_multiregion(e1, e2, threshold=5000):
"""
Consider multiregion if the distance between fragments is min >= threshold
"""
# sit on different chromosomes
if e1[CHR] != e2[CHR]:
return True
if e1[END] < e2[START] and e2[START] - e1[END] >= threshold:
return True
if e1[START] > e2[END] and e1[START] - e2[END] >= threshold:
return True
return False
def pick_chr(chrlen, conf):
"""
Choose ecDNA origin chromosome
"""
return random.choices([p for p in chrlen], weights=conf[WEIGHT_CHR], k=1)[0]
def pick_pos(chrranges, conf, chr_):
"""
Pick genomic position on chr
"""
return random.randrange(chrranges[chr_][0], chrranges[chr_][1], 5)
def pick_pos_overlap(ecDNA, index):
"""
Pick genomic position that overlaps the previous fragment (start,stop).
Args:
ecDNA (list): List of fragments
index (int): Index of the previous neighboring element
"""
start = min(ecDNA[index][START], ecDNA[index][END])
end = max(ecDNA[index][START], ecDNA[index][END])
return random.randrange(start, end, 5)
def pick_len(chrlen, conf, chr, start, meansize=300000, sd=5000):
"""
Pick length of ecDNA fragment
"""
return random.randrange(meansize - sd, meansize + sd, 100)
def pick_len_overlap(ecDNA, previous_index):
"""
Pick length of ecDNA fragment
"""
previous_len = abs(ecDNA[previous_index][START] - ecDNA[previous_index][END])
return random.randrange(int(previous_len / 4), int(previous_len / 2), 100)
def choose_fragment_random(chrlen, chrranges, conf):
"""
Choose genomic fragment of origin
"""
chr_ = pick_chr(chrlen, conf)
start_ = pick_pos(chrranges, conf, chr_)
len_ = pick_len(chrlen, chrranges, chr_, start_)
end_ = start_ + len_
return chr_, start_, end_, len_
def choose_fragment_neighbor(chrlen, chrranges, conf, index, ecDNA):
"""
Choose genomic fragment of origin
"""
if index == -1:
chr_ = pick_chr(chrlen, conf)
start_ = pick_pos(chrranges, conf, chr_)
len_ = pick_len(chrlen, conf, chr_, start_)
end_ = start_ + len_
else:
# 0 1 2 3 4 5
# chr_,start_,end_,strand_,count_frag
chr_ = ecDNA[index][0]
start_ = ecDNA[index][2]
len_ = pick_len(chrlen, conf, chr_, start_)
end_ = start_ + len_
return chr_, start_, end_, len_
def choose_fragment_pseudorandom(chrlen, chrranges, conf, index, ecDNA):
"""
Choose pseudorandom only with a probability is random
"""
neighbor = random.choices([1, 0], k=1)[0]
# choose to pick a neighboring fragment
if neighbor == 1:
chr_, start_, end_, len_ = choose_fragment_neighbor(chrlen, chrranges, conf, index, ecDNA)
else:
# choose random position on the same chromosome
if conf[SINGLE_CHR]:
# initialize chromosome
chr_ = pick_chr(chrlen, conf) if index == -1 else ecDNA[index][0]
else:
# pick random chromosome
chr_ = pick_chr(chrlen, conf)
start_ = pick_pos(chrranges, conf, chr_)
len_ = pick_len(chrlen, conf, chr_, start_)
end_ = start_ + len_
return chr_, start_, end_, len_
def choose_fragment(chrlen, chrranges, index, conf, ecDNA, topology, type="random"):
if conf[NEIGHBOR]:
chr_, start_, end_, len_ = choose_fragment_neighbor(chrlen, chrranges, conf, index, ecDNA)
elif type == "pseudorandom":
chr_, start_, end_, len_ = choose_fragment_pseudorandom(chrlen, chrranges, conf, index, ecDNA)
else:
chr_, start_, end_, len_ = choose_fragment_random(chrlen, chrranges, conf)
return chr_, start_, end_, len_
def choose_fold(chrlen, chrranges, previous_index, conf, ecDNA, topology):
"""
Choose foldback fragment
"""
chr_ = ecDNA[previous_index][0]
start_ = pick_pos_overlap(ecDNA, previous_index)
len_ = pick_len_overlap(ecDNA, previous_index)
end_ = start_ + len_
return chr_, start_, end_, len_
def crop_left_end(fragment_len, p):
"""
Decide to crop (10% of fragment len) or not the left end
"""
del_left = random.choices([1, 0], weights=[p, 100 - p], k=1)[0]
return del_left, 0.1 * fragment_len
def crop_right_end(fragment_len, p):
"""
Decide to crop (10% of fragment len) or not the right end
"""
del_right = random.choices([1, 0], weights=[p, 100 - p], k=1)[0]
return del_right, 0.1 * fragment_len
def small_deletions(chrlen, chr_, start_, end_, len_, conf, topology):
"""
Emulate small deletions
"""
(left, lsize) = crop_left_end(len_, conf[P_DEL_LEFT])
(right, rsize) = crop_right_end(len_, conf[P_DEL_RIGHT])
# crop left part
if left == 1:
start_ = min(chrlen[chr_], start_ + lsize)
topology[SMALL_DEL] += 1
# crop right part
if right == 1:
end_ = max(0, end_ - rsize)
topology[SMALL_DEL] += 1
return int(start_), int(end_)
def invert_event(conf):
"""
Invert fragment with a probability conf[P_INVERT]
"""
p = conf[P_INVERT]
return random.choices([1, 0], weights=[p, 100 - p], k=1)[0]
def duplicate_event(conf):
"""
Duplicate fragment with a probability conf[P_DUP]
"""
p = conf[P_DUP]
return random.choices([1, 0], weights=[p, 100 - p], k=1)[0]
def invert(conf, topology):
strand = "+"
if invert_event(conf) == 1:
topology[INV] += 1
strand = "-"
return strand
def duplicate(conf, topology):
if duplicate_event(conf) == 1:
topology[DUP] += 1
return True
return False
def pick_nr_copies(conf):
"""
Pick number of ecDNA copies
"""
return random.choices(range(400, 700, 10), k=1)[0]
def foldback(conf, topology):
"""
Foldback fragment with a probability conf[P_FOLDBACK]
"""
p = conf[P_FOLDBACK]
return random.choices([1, 0], weights=[p, 100 - p], k=1)[0]
def back(conf, topology):
"""
Backstructure conf[P_RETURN]
"""
p = conf[P_FOLDBACK]
return random.choices([1, 0], weights=[p, 100 - p], k=1)[0]
def choose(conf):
"""
0 - Random position
1 - Foldback
2 - Return
"""
p = conf[P_CHOOSE_RANDOM]
return random.choices([0, 1, 2], weights=[p, 3 * (100 - p) / 4, (100 - p) / 4], k=1)[0]
def compare_metainformation(acceptance_criteria, metastring):
for key in metastring:
if key in EVENTS:
id = EVENTS[key]
if acceptance_criteria[id] == 1 and metastring[key] == 0:
return False
return True
def simulation(chrlen, chrranges, conf, type="random", acceptance_criteria=[3, 0, 1, 1, 0, 1, 1, 1, 0]):
"""
Simulate a single ecDNA structure
"""
# 1. add all default parameters (unchanged) to conf file
for key in CONF_INIT:
if key not in conf:
conf[key] = CONF_INIT[key]
# pp.pprint(conf)
pass_structure = False
max_iterations = 300
count_iter = 0
while (not pass_structure) and count_iter < max_iterations:
ecDNA = []
count_frag = -1
count_iter += 1
topology = {N_FRAG: 0,
FRAG_LEN: 0,
SMALL_DEL: 0,
DUP: 0,
INV: 0,
INTERCHR: 0,
MULTI_REGION: 0,
FOLDBACK: 0,
RETURN: 0
}
# 2. start simulation
while len(ecDNA) < conf[N]:
eventtype = choose(conf)
if eventtype != 0 and count_frag == -1:
continue
if eventtype == 0:
# 2.1. choose fragment
chr_, start_, end_, len_ = choose_fragment(chrlen, chrranges, count_frag, conf, ecDNA, topology,
type=type)
# print("Choose fragment Nr. chr start end len, ", count_frag, chr_, start_, end_, len_)
elif eventtype == 1:
# foldback
topology[FOLDBACK] += 1
chr_, start_, end_, len_ = choose_fold(chrlen, chrranges, count_frag, conf, ecDNA, topology)
else:
# return/back
pass
# 2.2 emulate small deletions
start_, end_ = small_deletions(chrlen, chr_, start_, end_, len_, conf, topology)
# print("Small del Nr. chr start end len, ", count_frag, chr_, start_, end_, len_)
# 2.3 emulate inversion
strand_ = invert(conf, topology)
count_frag += 1
topology[N_FRAG] += 1
ecDNA.append((chr_, start_, end_, strand_, count_frag))
# mark topology as interchromosomal
if ecDNA[count_frag - 1][0] != ecDNA[count_frag][0]:
topology[INTERCHR] += 1
# multiregion
if is_multiregion(ecDNA[count_frag - 1], ecDNA[count_frag]):
topology[MULTI_REGION] += 1
# 2.4 allow simple duplication (1>kbp)
if abs(start_ - end_) > conf[MIN_TANDEM_DUPLICATION]:
while ((count_frag + 1) < conf[N]) and duplicate(conf, topology):
# if count_frag < conf[N]:
# dup_ = duplicate(conf)
# if dup_ == 1:
count_frag += 1
topology[N_FRAG] += 1
ecDNA.append((chr_, start_, end_, strand_, count_frag))
# 3. choose coverage
cov_ = pick_nr_copies(conf)
# 4. check if conformation correct
pass_structure = compare_metainformation(acceptance_criteria, topology)
# if pass_structure:
# print("Pass")
# print(acceptance_criteria)
# print(topology)
# print(ecDNA)
# print()
# else:
# print("No pass")
# print(ecDNA)
# print(topology)
# print()
if count_iter == max_iterations:
return None, None, topology
return ecDNA, cov_, topology
def ecDNA2bed(ecDNAdict, fout):
# print(os.getcwd())
# print(os.listdir())
with open(fout, "w") as f:
f.write("#chr\tstart\tstop\tdirection\ttarget\tcoverage\tstructure\tfragment\n")
for circ in ecDNAdict:
for i, frag in enumerate(ecDNAdict[circ]["structure"]):
chr_, start_, end_, strand_, count_frag = frag
strand_ = "+" if strand_ == 0 else "-"
coverage_ = ecDNAdict[circ]["coverage"]
circ_ = """circ_{}""".format(str(circ))
circ_ = """circ_{}""".format(str(circ))
f.write("""{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n""".format(chr_,
str(start_),
str(end_),
strand_,
circ_,
str(coverage_),
circ_,
str(i)
))
def ecDNA2bed_single(ecDNAdict, fout):
# print(os.getcwd())
# print(os.listdir())
os.makedirs(os.path.dirname(fout), exist_ok=True)
with open(fout, "w") as f:
f.write("#chr\tstart\tstop\tdirection\ttarget\tcoverage\tstructure\tfragment\n")
for frag in ecDNAdict["structure"]:
chr_, start_, end_, strand_, count_frag = frag
strand_ = strand_
coverage_ = ecDNAdict["coverage"]
circ_ = """circ_0"""
f.write("""{}\t{}\t{}\t{}\t{}\t{}\t{}\t{}\n""".format(chr_,
str(start_),
str(end_),
strand_,
circ_,
str(coverage_),
circ_,
str(count_frag)
))