-
Notifications
You must be signed in to change notification settings - Fork 0
/
sim.py
1341 lines (1166 loc) · 52.4 KB
/
sim.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
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# file: sim.py
# author: Jingyi Wang and Xuecong Fu
# the file is originated from sim.py in TUSV codes by Jingyi, Xuecong fixed bugs and extend the simulation to include
# allelic inter/intra-chromosomal translocations and single nucleotide variants (SNVs)
# purpose: the main simulation code
##########
# Import #
##########
from __future__ import division
import copy
import sys
import os
import argparse
import vcf
import itertools
import random
import operator
import datetime
import shutil
import pickle
import pandas as pd
import numpy as np
import graphviz as gv
import chrm_prof as chpr
import gene_prof as gnpr
from chrm_prof import *
sys.path.insert(0, 'helper/')
import combine_copy_nums as ccn
#############
# Functions #
#############
def printnow(s, newline = True):
s = str(s)
if newline:
s += '\n'
sys.stdout.write(s)
sys.stdout.flush()
def main(argv):
args = get_args(argv)
# input arguments
m = args['m']
n = args['n']
num_mutes = args['num_mutes']
num_mutes_snv = args['num_mutes_snv']
num_patients = args['num_patients']
directory = os.path.dirname(os.path.realpath(__file__))
# for num_mutes in [10, 50, 100]:
# for m in [1, 3, 5, 10]:
# for n in [2, 3, 4, 5]:
# print 'n:', n, 'm:', m, 'num_mutes:', num_mutes
size_mutes = args['size_mutes']
metaFile = args['meta_file']
output_folder = args['output_folder']
constants_dict = dict()
constants_dict['mut_types'] = ['amp','amp', 'inv', 'rem', 'rem', 'trans']
#constants_dict['mut_types'] = [ 'rem']
constants_dict['exp_mut_size'] = size_mutes # default exp_mut_size is 5745000
constants_dict['exp_mut_count'] = num_mutes / ( 2 * n - 2)
if num_mutes_snv is not None:
constants_dict['snv_mut_lambda'] = num_mutes_snv / (2*n - 2)
else:
constants_dict['snv_mut_lambda'] = None
constants_dict['cov'] = 20
constants_dict['read_len'] = 300
constants_dict['num_patients'] = num_patients
constants_dict['num_leaves'] = n
constants_dict['num_samples'] = m
constants_dict['deterministic']=args["deterministic"]
constants_dict['read_depth'] = args["read_depth"]
constants_dict['only_leaf'] = args['only_leaf']
# remove chrom_dict later
chrom_dict = dict()
# chrom_dict[('1', 0)] = chpr.ChrmProf(10000000, '1', 0) # from ICGC CNV data
# chrom_dict[('1', 1)] = chpr.ChrmProf(10000000, '1', 1)
# chrom_dict[('2', 0)] = chpr.ChrmProf(10000000, '1', 0) # from ICGC CNV data
# chrom_dict[('2', 1)] = chpr.ChrmProf(10000000, '1', 1)
chrom_dict[('1', 0)] = chpr.ChrmProf(249198692, '1', 0) #from ICGC CNV data
chrom_dict[('1', 1)] = chpr.ChrmProf(249198692, '1', 1)
chrom_dict[('2', 0)] = chpr.ChrmProf(243048760, '2', 0)
chrom_dict[('2', 1)] = chpr.ChrmProf(243048760, '2', 1)
# chrom_dict[('3', 0)] = chpr.ChrmProf(197856433, '3', 0)
# chrom_dict[('3', 1)] = chpr.ChrmProf(197856433, '3', 1)
# chrom_dict[('4', 0)] = chpr.ChrmProf(190921709, '4', 0)
# chrom_dict[('4', 1)] = chpr.ChrmProf(190921709, '4', 1)
# chrom_dict[('5', 0)] = chpr.ChrmProf(186092833, '5', 0)
# chrom_dict[('5', 1)] = chpr.ChrmProf(186092833, '5', 1)
# chrom_dict[('6', 0)] = chpr.ChrmProf(170918031, '6', 0)
# chrom_dict[('6', 1)] = chpr.ChrmProf(170918031, '6', 1)
# chrom_dict[('7', 0)] = chpr.ChrmProf(159119220, '7', 0)
# chrom_dict[('7', 1)] = chpr.ChrmProf(159119220, '7', 1)
# chrom_dict[('8', 0)] = chpr.ChrmProf(146293414, '8', 0)
# chrom_dict[('8', 1)] = chpr.ChrmProf(146293414, '8', 1)
# chrom_dict[('9', 0)] = chpr.ChrmProf(141071475, '9', 0)
# chrom_dict[('9', 1)] = chpr.ChrmProf(141071475, '9', 1)
# chrom_dict[('10', 0)] = chpr.ChrmProf(135434551, '10', 0)
# chrom_dict[('10', 1)] = chpr.ChrmProf(135434551, '10', 1)
# chrom_dict[('11', 0)] = chpr.ChrmProf(134944770, '11', 0)
# chrom_dict[('11', 1)] = chpr.ChrmProf(134944770, '11', 1)
# chrom_dict[('12', 0)] = chpr.ChrmProf(133777645, '12', 0)
# chrom_dict[('12', 1)] = chpr.ChrmProf(133777645, '12', 1)
# chrom_dict[('13', 0)] = chpr.ChrmProf(115106996, '13', 0)
# chrom_dict[('13', 1)] = chpr.ChrmProf(115106996, '13', 1)
# chrom_dict[('14', 0)] = chpr.ChrmProf(107285437, '14', 0)
# chrom_dict[('14', 1)] = chpr.ChrmProf(107285437, '14', 1)
# chrom_dict[('15', 0)] = chpr.ChrmProf(102400037, '15', 0)
# chrom_dict[('15', 1)] = chpr.ChrmProf(102400037, '15', 1)
# chrom_dict[('16', 0)] = chpr.ChrmProf(90163275, '16', 0)
# chrom_dict[('16', 1)] = chpr.ChrmProf(90163275, '16', 1)
# chrom_dict[('17', 0)] = chpr.ChrmProf(81048659, '17', 0)
# chrom_dict[('17', 1)] = chpr.ChrmProf(81048659, '17', 1)
# chrom_dict[('18', 0)] = chpr.ChrmProf(78015057, '18', 0)
# chrom_dict[('18', 1)] = chpr.ChrmProf(78015057, '18', 1)
# chrom_dict[('19', 0)] = chpr.ChrmProf(59095126, '19', 0)
# chrom_dict[('19', 1)] = chpr.ChrmProf(59095126, '19', 1)
# chrom_dict[('20', 0)] = chpr.ChrmProf(62912463, '20', 0)
# chrom_dict[('20', 1)] = chpr.ChrmProf(62912463, '20', 1)
# chrom_dict[('21', 0)] = chpr.ChrmProf(48084820, '21', 0)
# chrom_dict[('21', 1)] = chpr.ChrmProf(48084820, '21', 1)
# chrom_dict[('22', 0)] = chpr.ChrmProf(51219006, '22', 0)
# chrom_dict[('22', 1)] = chpr.ChrmProf(51219006, '22', 1)
# chrom_dict[('23', 0)] = chpr.ChrmProf(155233846, '23', 0)
# chrom_dict[('23', 1)] = chpr.ChrmProf(155233846, '23', 1)
# sub_folder_name = 'n_' + str(n) + '_m_' + str(m) + '_l_' + str(num_mutes)
if not os.path.exists(output_folder):
os.mkdir(output_folder)
readme = open(output_folder + "/README.md", 'w')
for key, value in constants_dict.items():
readme.write(str(key) + ":" + str(value) + "\n")
readme.close()
for patient_idx in range(1, 1 + constants_dict['num_patients']):
patient_folder_name = 'patient' + str(patient_idx)
# outputFolder = directory + '/sim_data' + '/' + sub_folder_name + '/' + patient_folder_name
outputFolder = output_folder + '/' + patient_folder_name
# clean up existing files under outputFolder
if os.path.exists(outputFolder):
shutil.rmtree(outputFolder)
os.makedirs(outputFolder)
l = random_get_tree(n) # list
#print(l)
edge_list = get_edges(l) ###xf: generate edges list with format of [(0,1,'r'/'l'),...]
#print(edge_list)
gp = gnpr.GeneProf(chrom_dict, constants_dict)
t = Tree(edge_list, gp)
#print(t.node_list, t.idx_node_dict)
geneprof_list = list()
t.add_mutations_along_edges(t.rootNode, geneprof_list)
generate_t(t, 'T.dot', outputFolder)
#U = np.array([[1/(2*n-1)]*(2*n-1)])
if not constants_dict['only_leaf']:
U = random_get_usages(m, 2 * n - 1)
else:
U_original = random_get_usages(m, n + 1)
U = np.zeros((m, 2 * n -1))
U[:, :n] = U_original[:, :n]
U[:, -1] = U_original[:, -1]
print('U', U)
l, sv_cn_idx_dict = get_bp_copy_num_idx_dict(t, n, constants_dict)
r, seg_cn_idx_dict, seg_bgn_idx_dict, seg_end_idx_dict = get_seg_copy_num_idx_dict(t, n)
with open(outputFolder + "/dimension", 'w') as f:
f.write("(" + str(l) + "," + str(r) + ")")
### xf: combine the segment settings from both two alleles and also different node from mutations
bool_list = np.random.choice([True, False], r)
if constants_dict['snv_mut_lambda'] is None:
C = generate_c(t, n, constants_dict, bool_list)
c_p, c_m = generate_seg_cp_paternal(t, n, bool_list)
###xf: need to be further editted
W, W_SV, _ = generate_w(C, t.idx_node_dict, l, 0)
F, df_list = generate_f(U, C, l, 0, r, seg_cn_idx_dict, sv_cn_idx_dict, None, constants_dict['deterministic'], None,None,None, None)
a, h, mate_dict = get_a_h_mate_dict(t, n, constants_dict)
generate_s(metaFile, t, l, sv_cn_idx_dict, r, seg_cn_idx_dict, seg_bgn_idx_dict, seg_end_idx_dict, F, U, C,
c_p, c_m, a, h, mate_dict, outputFolder)
output_tsv(U, '/U.tsv', outputFolder)
output_tsv(C, '/C.tsv', outputFolder)
output_tsv(W, '/W.tsv', outputFolder)
output_tsv(W_SV, '/W_SV.tsv', outputFolder)
output_tsv(F, '/F.tsv', outputFolder)
else:
N = np.random.poisson(constants_dict['read_depth'], (m, r))
g, snv_cn_idx_dict = get_snv_copy_num_idx_dict(t)
#print(snv_cn_idx_dict)
C_list, C_unsampled_snv_list, snv_sampled_idx_list, snv_unsampled_idx_list, d_list, d_unsampled_list = \
generate_c_snv(t, n, constants_dict, bool_list, [1])
c_p, c_m = generate_seg_cp_paternal(t, n, bool_list)
F_list = []
F_unsampled_snv_list = []
a, h, mate_dict = get_a_h_mate_dict(t, n, constants_dict)
output_tsv(U, '/U.tsv', outputFolder)
i=0
C = C_list[i]
#print(t.idx_node_dict)
B, B_SV, B_SNV = generate_b(C, l, g)
W, W_SV, W_SNV = generate_w(C, t.idx_node_dict, l, g)
F, df_list, alt_counts_sv, alt_counts_snv, ref_counts_sv, ref_counts_snv, phylowgs_cnv_info = generate_f(U, C_list[i], l, len(snv_sampled_idx_list[i]), r, seg_cn_idx_dict, sv_cn_idx_dict,
snv_cn_idx_dict, constants_dict['deterministic'], N, snv_sampled_idx_list[i], d_list[i],
bool_list, constants_dict['read_depth'])
F_list.append(F)
#print(F[:, (l+len(snv_sampled_idx_list[i])):])
F_unsampled_snv = generate_f_unsampled(U, C_unsampled_snv_list[i],
C_list[i][:, (l + len(snv_sampled_idx_list[i])):], r,
seg_cn_idx_dict, snv_cn_idx_dict,
constants_dict['deterministic'], snv_unsampled_idx_list[i],
d_unsampled_list[i], bool_list, N,
F[:, (l + len(snv_sampled_idx_list[i])):])
F_unsampled_snv_list.append(F_unsampled_snv)
cntmd_dict, df_phylowgs_list = generate_s_snv(metaFile, t, l, sv_cn_idx_dict, r, seg_cn_idx_dict, g, snv_cn_idx_dict,
snv_sampled_idx_list[i], snv_unsampled_idx_list[i], seg_bgn_idx_dict, seg_end_idx_dict, F,
F_unsampled_snv, U, C_list[i], c_p, c_m, a, h, mate_dict, outputFolder, "", alt_counts_sv,
alt_counts_snv, ref_counts_sv, ref_counts_snv, phylowgs_cnv_info)
if not os.path.exists(outputFolder + '/fastclone_input'):
os.mkdir(outputFolder + '/fastclone_input')
if not os.path.exists(outputFolder + '/pyclone_input'):
os.mkdir(outputFolder + '/pyclone_input')
if not os.path.exists(outputFolder + '/cntmd_input'):
os.mkdir(outputFolder + '/cntmd_input')
with open('cntmd_dict.pickle', 'wb') as f:
pickle.dump(cntmd_dict, f)
for i in range(len(df_list)):
df_list[i].to_csv(outputFolder + '/fastclone_input/sample' + str(i) + '.tsv', sep='\t')
df_list[i]["minor_cn"] = np.round(df_list[i]["minor_cn"]).astype(int)
df_list[i]["major_cn"] = np.round(df_list[i]["major_cn"]).astype(int)
df_list[i].to_csv(outputFolder + '/pyclone_input/sample' + str(i) + '.tsv', sep='\t')
for i in range(len(df_phylowgs_list)):
df_phylowgs_list[i].to_csv(outputFolder + '/phylowgs/cnv' + str(i) + '.tsv', sep='\t', index=False)
output_tsv(C, '/C.tsv', outputFolder)
output_tsv(W, '/W.tsv', outputFolder)
output_tsv(W_SV, '/W_SV.tsv', outputFolder)
output_tsv(W_SNV, '/W_SNV.tsv', outputFolder)
output_tsv(B, '/B.tsv', outputFolder)
output_tsv(B_SV, '/B_SV.tsv', outputFolder)
output_tsv(B_SNV, '/B_SNV.tsv', outputFolder)
output_tsv(F, '/F.tsv', outputFolder)
# output_tsv(F_unsampled_snv, '/F_unsampled_snv.tsv', outputFolder)
# output_tsv(C_unsampled_snv_list[i], '/C_unsampled_snv.tsv', outputFolder)
edge_list_pickle = open(outputFolder + "/edge_list.pickle", 'wb')
pickle.dump(edge_list, edge_list_pickle)
edge_list_pickle.close()
# given a number n, generate all possible directed binary trees with n nodes.
# eg. if n = 4, return [ [1,[1,[1,1]]], [1,[[1,1],1]], [[1,1],[1,1]], [[1,[1,1]],1], [[[1,1],1],1] ]
# each 1 in the list represents a tree node.
def all_possible_trees(n):
l = list()
for tree in all_possible_trees_helper(n):
l.append(tree)
return l
def all_possible_trees_helper(n):
# base case
if n == 1:
yield 1
# recursive case
for i in range(1, n):
left_list = all_possible_trees_helper(i)
right_list = all_possible_trees_helper(n - i)
for left, right in itertools.product(left_list, right_list):
yield [left, right]
# input n (number of leaves), output a random tree represented by a list.
# tree node is represented by 1.
def random_get_tree(n): ###xf: select a random tree from all possible trees
l = all_possible_trees(n)
idx = random.randint(0, len(l) - 1)
return l[idx]
# input a tree (list), return number of leaves
def get_number_of_leaves(l):
if l[0] == 1 and l[1] == 1:
return 2
elif l[0] == 1 and l[1] != 1:
return 1 + get_number_of_leaves(l[1])
elif [0] != 1 and l[1] == 1:
return 1 + get_number_of_leaves(l[0])
else:
return get_number_of_leaves(l[0]) + get_number_of_leaves(l[1])
# given a tree (list), return list of edges(tuple). tuple format: (id, parent_id, l/r)
# l/r: left or right child
def get_edges(l):
n = get_number_of_leaves(l)
p = 2 * n - 1 # idx of root node
leaf_list = list(range(1, n + 1)) # indices of leaf nodes
steiner_list = list(range(n, p, 1)) ### xf: debug the tree generation part by adding a fixed steiner node list
result = list()
get_edges_helper(l, n, leaf_list, p, result, steiner_list)
return result
def get_edges_helper(l, n, leaf_list, p, result, steiner_node_list): ### xf: debug the tree generation part by adding a fixed steiner node list
left, right = l[0], l[1]
if left == 1 and right == 1:
result.append((leaf_list[0], p, 'l'))
result.append((leaf_list[1], p, 'r'))
elif left == 1 and right != 1:
result.append((leaf_list[0], p, 'l'))
steiner_node = steiner_node_list.pop()
result.append((steiner_node, p, 'r'))
get_edges_helper(l[1], n-1, leaf_list[1:], steiner_node, result, steiner_node_list)
elif right == 1 and left != 1:
result.append((leaf_list[-1], p, 'r'))
steiner_node = steiner_node_list.pop()
result.append((steiner_node, p, 'l'))
get_edges_helper(l[0], n-1, leaf_list[:-1], steiner_node, result, steiner_node_list)
else:
n_left = get_number_of_leaves(l[0])
leaf_list_left = leaf_list[:n_left]
leaf_list_right = leaf_list[n_left:]
steiner_node1 = steiner_node_list.pop()
steiner_node2 = steiner_node_list.pop()
result.append((steiner_node1, p, 'l')) # left
result.append((steiner_node2, p, 'r')) # right
get_edges_helper(l[0], n_left, leaf_list_left, steiner_node1, result, steiner_node_list) # left
get_edges_helper(l[1], n - n_left, leaf_list_right, steiner_node2, result, steiner_node_list) # right
return result, steiner_node_list
# pwd: os.path.dirname(os.path.realpath(__file__))
# generate a dot file to save the random tree with n nodes and random mutations
def generate_t(tree, filename, directory):
dot = gv.Digraph()
nodes = tree.node_list
edges = get_graph_edges(tree.edge_list)
add_nodes(dot, nodes)
add_edges(dot, edges)
# print dot.source
dot.save(filename, directory)
return
def add_nodes(graph, nodes):
for n in nodes:
if isinstance(n, tuple):
graph.node(str(n[0]), **n[1])
else:
graph.node(str(n))
return graph
def add_edges(graph, edges):
for e in edges:
if isinstance(e[0], tuple):
graph.edge(*e[0], **e[1])
else:
graph.edge(*e)
return graph
# edge_list: list of (node_id (int), parent_node_id (int)) tuples.
# return list of (from (str), to (str)) tuples.
def get_graph_edges(edge_list):
result = list()
for(node_id, parent_node_id, lr) in edge_list:
result.append((str(parent_node_id), str(node_id)))
return result
# return a 1 by n np array sum to one
def get_usage_for_one_patient(n):
a = np.random.dirichlet(np.ones(n),size = 1)
return a
# return m by n np array, each row sum to one
def random_get_usages(m, n):
a = get_usage_for_one_patient(n)
for i in range(1, m):
b = get_usage_for_one_patient(n)
a = np.concatenate((a, b), axis = 0)
return a
### xf: add snv copy number idx dict
def get_snv_copy_num_idx_dict(tree):
d = set()
d2 = {}
for idx in tree.node_list:
temp_snv_dict = tree.idx_node_dict[idx].geneProf.get_snv_dict()
for tuple in temp_snv_dict.keys():
if tuple not in d:
d.add(tuple)
sorted_d = sorted(list(d),key=lambda x: (int(x[0]), x[1]))
for k in range(len(sorted_d)):
d2[sorted_d[k]] = k
#print(d2)
return len(list(d)), d2
# input a tree (Tree object) and n (number of leaf nodes)
# output a l (number of bps) and a dictionary
# key: chrom
# val: dictionary
# key: (pos, isLeft) tuple. When two bps have the same pos, the one with isLeft == True comes first.
# val: index of the bp
def get_bp_copy_num_idx_dict(tree, n, constants_dict):
# put all bps in leaf nodes into dictionary d1
d1 = dict()
d_chrom = {}
for idx in tree.node_list:
temp_bp_dict = tree.idx_node_dict[idx].geneProf.get_sv_read_nums_dict(constants_dict['cov'], constants_dict['read_len'])
###xf: for all chromosomes for both alleles
for chrom in temp_bp_dict.keys():
if chrom not in d1:
d1[chrom] = set()
for (pos, isLeft, chr_) in temp_bp_dict[chrom]:
if (pos, isLeft, chr_) not in d1[chrom]:
d1[chrom].add((pos, isLeft, chr_))
# sort vals in d1 based on pos and isLeft (isLeft == True comes first) and chrom
idx = 0
sorted_chrom = sorted(d1.keys(), key=int)
d2 = dict()
for chrom in sorted_chrom:
#print(d1[chrom])
d2[chrom] = dict()
# for i in range(0,len(d1[chrom])):
# for j in range(i+1,len(d1[chrom])):
# if list(d1[chrom])[i][0] == list(d1[chrom])[j][0]:
# print(list(d1[chrom])[i])
sorted_pos_list = sorted(list(set(map(operator.itemgetter(0), d1[chrom]))))
#print(len(sorted_pos_list))
for pos in sorted_pos_list:
if (pos, True, chrom) in d1[chrom]:
d2[chrom][(pos, True, chrom)] = idx
idx += 1
if (pos, False, chrom) in d1[chrom]:
d2[chrom][(pos, False, chrom)] = idx
idx += 1
#print("d1", chrom, len(d1[chrom]))
#print("d2",chrom, len(d2[chrom]))
return idx, d2
# input a tree (Tree object) and n (number of leaf nodes)
# output a r and a dictionary
# key: chrom
# val: dictionary
# key: (bgn, end) tuple
# val: index
def get_seg_copy_num_idx_dict(tree, n):
d1 = dict()
for idx in tree.node_list: ### xf: go through all the nodes in the tree
temp_copy_nums_dict = tree.idx_node_dict[idx].geneProf.get_copy_nums_dict()
for chrom in temp_copy_nums_dict.keys():
if chrom not in d1:
d1[chrom] = list()
(bgns, ends, cps1, cps2) = temp_copy_nums_dict[chrom]
d1[chrom].append([bgns, ends, cps1, cps2])
idx = 0
sorted_chrom = sorted(d1.keys(), key=int)
d2 = dict()
bgn2idx = dict()
end2idx = dict()
for chrom in sorted_chrom:
d2[chrom] = dict()
# random generate usages to match the input format
usages = [float(1/len(d1[chrom]))] * len(d1[chrom])
[res_bgns, res_ends, res_cps1, res_cps2] = ccn.combine_copy_nums_quartet(d1[chrom], usages)
for i in range(len(res_bgns)):
d2[chrom][(res_bgns[i], res_ends[i])] = idx
bgn2idx[(chrom, res_bgns[i])] = idx
end2idx[(chrom, res_ends[i])] = idx
idx += 1
return idx, d2, bgn2idx, end2idx
def make_2d_list(rows, cols):
result = list()
for r in range(rows):
result.append([0] * cols)
return result
def generate_b(C, l, g):
C_sv_snv = C[:, : l + g]
B = copy.deepcopy(C_sv_snv)
B[B>1] = 1
B_sv = B[:, :l]
B_snv = B[:, l:(l+g)]
return B, B_sv, B_snv
def generate_w(C, idx_node_dict, l, g):
C_sv_snv = C[:, : l+g]
W = np.zeros(C_sv_snv.shape)
for node_idx in idx_node_dict.keys():
if node_idx != C.shape[0]:
parent_idx = idx_node_dict[node_idx].parent.index
indices_a = np.where(C_sv_snv[node_idx-1] - C_sv_snv[parent_idx-1] > 0)[0]
indices_b = np.where(C_sv_snv[parent_idx-1] == 0)[0]
indices = np.intersect1d(indices_a, indices_b)
W[node_idx-1, indices] = 1
W_sv = W[:, :l]
if g != 0:
W_snv = W[:, l:(l+g)]
else:
W_snv = None
return W, W_sv, W_snv
def generate_c_snv(tree, n, constants_dict, bool_list, subsample_list=[0, 0.0005, 0.001, 0.002]):
l, sv_cn_idx_dict = get_bp_copy_num_idx_dict(tree, n, constants_dict)
r, seg_cn_idx_dict, seg_bgn_idx_dict, seg_end_idx_dict = get_seg_copy_num_idx_dict(tree, n)
g, snv_cn_idx_dict = get_snv_copy_num_idx_dict(tree)
c_list = []
c_unsampled_list = []
snv_sampled_idx_list = []
snv_unsampled_idx_list = []
g_subsample_list = []
d_list = []
d_unsampled_list = []
for subsample in subsample_list:
g_subsample_list.append(int(g*subsample))
for g_sample_idx in range(len(g_subsample_list)):
g_sample = g_subsample_list[g_sample_idx]
if g_sample_idx == 0:
snv_sampled_idx = np.sort(np.random.choice(g, size=g_sample, replace=False))
else:
snv_sampled_idx = np.sort(np.append(snv_sampled_idx, np.random.choice(snv_unsampled_idx, size=g_sample - g_subsample_list[g_sample_idx -1], replace=False)))
snv_unsampled_idx = np.setdiff1d(np.arange(g), snv_sampled_idx)
#print(snv_unsampled_idx)
c = make_2d_list(len(tree.node_list), (l + g_sample + 2*r))
d_sampled = np.zeros((l+g_sample))
c_unsampled_snv = make_2d_list(len(tree.node_list), (g-g_sample))
d_unsampled = np.zeros((g-g_sample))
for idx in tree.node_list:
print("node",idx)
row = idx - 1
# add copy number for break points
temp_bp_dict = tree.idx_node_dict[idx].geneProf.get_sv_read_nums_dict(constants_dict['cov'], constants_dict['read_len'])
for chrom in temp_bp_dict:
for (pos, isLeft, chr_,) in temp_bp_dict[chrom]:
cp = temp_bp_dict[chrom][(pos, isLeft, chr_)]["copy_num"]
d = temp_bp_dict[chrom][(pos, isLeft, chr_)]["pm"]
col = sv_cn_idx_dict[chrom][(pos, isLeft, chr_)]
c[row][col] = cp
d_sampled[col] = d
# add copy number for segments
temp_copy_nums_dict = tree.idx_node_dict[idx].geneProf.get_copy_nums_dict()
#print(idx, "temp_copy_nums_dict", temp_copy_nums_dict)
for chrom in temp_copy_nums_dict:
(bgns, ends, cps1, cps2) = temp_copy_nums_dict[chrom]
for i in range(len(bgns)):
cp1 = cps1[i]
cp2 = cps2[i]
seg_indices_list = get_indices_for_segment(seg_bgn_idx_dict, seg_end_idx_dict, (chrom, bgns[i]), (chrom, ends[i]))
for j in range(len(seg_indices_list)):
col = seg_indices_list[j] + l + g_sample
if bool_list[col-l-g_sample]:
c[row][col] = cp1
c[row][col + r] = cp2
else:
c[row][col] = cp2
c[row][col + r] = cp1
temp_snv_dict = tree.idx_node_dict[idx].geneProf.get_snv_dict()
#print(idx, 'temp_snv_dict', temp_snv_dict)
#print(temp_copy_nums_dict)
print(temp_copy_nums_dict, temp_snv_dict)
for (chrm, pos) in temp_snv_dict.keys():
if snv_cn_idx_dict[(chrm, pos)] in snv_sampled_idx:
cn_idx = search_sv_cnv_num(chrm, pos, seg_cn_idx_dict)
cp = temp_snv_dict[(chrm, pos)]["copy_num"]
d = temp_snv_dict[(chrm, pos)]["pm"]
if bool_list[cn_idx]:
if cp > c[row][l+g+cn_idx + r*d]:
print("snv ", chrm, pos, "cp:", cp, "corres cnv cp:", c[row][l+g+cn_idx + r*d], c[row][l+g+cn_idx + abs(1-d)*r])
else:
if cp > c[row][l+g+cn_idx + r*abs(d-1)]:
print("snv ", chrm, pos, "cp:", cp, "corres cnv cp:", c[row][l+g+cn_idx + r*abs(d-1)], c[row][l+g+cn_idx + r*abs(d)])
col = np.where(snv_sampled_idx == snv_cn_idx_dict[(chrm, pos)])[0][0] + l
c[row][col] = cp
d_sampled[col] = d
else:
cn_idx = search_sv_cnv_num(chrm, pos, seg_cn_idx_dict)
cp = temp_snv_dict[(chrm, pos)]["copy_num"]
d = temp_snv_dict[(chrm, pos)]["pm"]
print("snv ", chrm, pos, "cp:", cp, "corres cnv cp:", c[row][l + g + cn_idx + r * d])
col = np.where(snv_unsampled_idx == snv_cn_idx_dict[(chrm, pos)])[0][0]
c_unsampled_snv[row][col] = cp
d_unsampled[col] = d
c_list.append(np.array(c))
c_unsampled_list.append(np.array(c_unsampled_snv))
snv_sampled_idx_list.append(snv_sampled_idx)
snv_unsampled_idx_list.append(snv_unsampled_idx)
d_list.append(d_sampled)
d_unsampled_list.append(d_unsampled)
return c_list, c_unsampled_list, snv_sampled_idx_list, snv_unsampled_idx_list, d_list, d_unsampled_list
# loop through each node in tree(Tree),
# for each treeNode: use self.get_copy_nums_dict() to get bgns, ends, cps list for each chromosomes
# use self.get_sv_read_nums_dict(cov, read_len) to get bps and their corresponding information for each chromosomes
# output c ((2n-1)*(l+r) matrix) ### xf: --> (2n-1)*(l+2r)
def generate_c(tree, n, constants_dict, bool_list):
l, sv_cn_idx_dict = get_bp_copy_num_idx_dict(tree, n, constants_dict)
r, seg_cn_idx_dict, seg_bgn_idx_dict, seg_end_idx_dict = get_seg_copy_num_idx_dict(tree, n)
c = make_2d_list(len(tree.node_list), (l + 2*r))
for idx in tree.node_list:
row = idx - 1
# add copy number for break points
temp_bp_dict = tree.idx_node_dict[idx].geneProf.get_sv_read_nums_dict(constants_dict['cov'], constants_dict['read_len'])
for chrom in temp_bp_dict:
for (pos, isLeft, chr_,) in temp_bp_dict[chrom]:
cp = temp_bp_dict[chrom][(pos, isLeft, chr_)]["copy_num"]
col = sv_cn_idx_dict[chrom][(pos, isLeft, chr_)]
c[row][col] = cp
# add copy number for segments
temp_copy_nums_dict = tree.idx_node_dict[idx].geneProf.get_copy_nums_dict()
for chrom in temp_copy_nums_dict:
(bgns, ends, cps1, cps2) = temp_copy_nums_dict[chrom]
for i in range(len(bgns)):
cp1 = cps1[i]
cp2 = cps2[i]
seg_indices_list = get_indices_for_segment(seg_bgn_idx_dict, seg_end_idx_dict, (chrom, bgns[i]), (chrom, ends[i]))
for j in range(len(seg_indices_list)):
col = seg_indices_list[j] + l
if bool_list[col - l]:
c[row][col] = cp1
c[row][col + r] = cp2
else:
c[row][col] = cp2
c[row][col + r] = cp1
result = np.array(c)
return result
# given start and end position, output list of segment indices (continuous)
# s = (chrom, bgn_pos), e = (chrom, end_pos)
def get_indices_for_segment(bgn2idx, end2idx, s, e):
result = list()
firstIdx = bgn2idx[s]
lastIdx = end2idx[e]
for i in range(firstIdx, lastIdx + 1):
result.append(i)
return result
# return a ((2n-1) * r) matrix contains paternal chrom copy number for each segment
# and a ((2n-1) * r) matrix contains maternal chrom copy number for each segment
def generate_seg_cp_paternal(tree, n, bool_list):
r, seg_cn_idx_dict, seg_bgn_idx_dict, seg_end_idx_dict = get_seg_copy_num_idx_dict(tree, n)
c_p = make_2d_list(len(tree.node_list), r)
c_m = make_2d_list(len(tree.node_list), r)
for idx in tree.node_list:
row = idx - 1
temp_chrom_dict = tree.idx_node_dict[idx].geneProf.chrom_dict
# temp_copy_nums_dict = tree.idx_node_dict[idx].geneProf.get_copy_nums_dict()
for chrom in list(filter(lambda x: x[1] == 0, temp_chrom_dict.keys())):
(bgns, ends, cps) = temp_chrom_dict[chrom].get_copy_nums()
for i in range(len(bgns)):
cp = cps[i]
seg_indices_list = get_indices_for_segment(seg_bgn_idx_dict, seg_end_idx_dict, (chrom[0], bgns[i]), (chrom[0], ends[i]))
for col in seg_indices_list:
if bool_list[col]:
c_p[row][col] = cp
else:
c_m[row][col] = cp
for chrom in list(filter(lambda x: x[1] == 1, temp_chrom_dict.keys())):
(bgns, ends, cps) = temp_chrom_dict[chrom].get_copy_nums()
for i in range(len(bgns)):
cp = cps[i]
seg_indices_list = get_indices_for_segment(seg_bgn_idx_dict, seg_end_idx_dict, (chrom[0], bgns[i]), (chrom[0], ends[i]))
for col in seg_indices_list:
if bool_list[col]:
c_m[row][col] = cp
else:
c_p[row][col] = cp
return c_p, c_m
# given u (m * (2n-1) matrix) and c ((2n-1)*(l+r) matrix), output f (m * (l+r) matrix)
### xf: deterministic f vs. generative f
def generate_f(u, c, l, g, r, seg_cn_idx_dict, sv_cn_idx_dict, snv_cn_idx_dict, det, N, snv_idx_list, d, bool_list, read_depth):
#print(u.shape, c.shape)
#print(seg_cn_idx_dict, sv_cn_idx_dict, snv_cn_idx_dict, det, N, snv_idx_list, d, bool_list)
print(l, g, r, c.shape)
print(d)
print(c)
if det:
return np.dot(u, c)
else:
df_pyclone_list = []
for i in range(u.shape[0]):
df_pyclone = pd.DataFrame(columns=['mutation_id','ref_counts','var_counts','normal_cn','minor_cn','major_cn'])
df_pyclone_list.append(df_pyclone)
F_true = np.dot(u, c)
print(c[:,l+g:])
F_gen = np.zeros((F_true.shape))
m = F_true.shape[0]
A_sv, R_sv, A_snv,R_snv = np.zeros((m, l)), np.zeros((m, l)), np.zeros((m, g)), np.zeros((m, g))
N = np.random.poisson(read_depth * F_true[:, l+g:])
sv_tuple_list = []
#print("seg_cn_idx_dict", seg_cn_idx_dict)
for chrom in sv_cn_idx_dict.keys():
for pos_tuple in sv_cn_idx_dict[chrom].keys():
sv_tuple_list.append((chrom, pos_tuple[0],sv_cn_idx_dict[chrom][pos_tuple]))
sv_tuple_list = sorted(sv_tuple_list, key=lambda a: a[2])
snv_tuple_list = list(snv_cn_idx_dict.items())
n, _ = c.shape
### generate F for CNVs with randomness
phylowgs_cnv_info = np.zeros((m, r, 3))
for i in range(0, r):
#print(F_true[:, l+g+i]/(F_true[:, l+g+i] + F_true[:, l+g+i+r]))
F_gen[:, l+g+i] = N[:, i]/read_depth
F_gen[:, l+g+r+i] = N[:,i+r]/read_depth
print(i, c[:,l+g+i], c[:,l+g+r+i])
if np.all(c[:,l+g+i] == 1) and np.all(c[:,l+g+r+i] == 1):
print("1")
phylowgs_cnv_info[:, i, 0] = 1
phylowgs_cnv_info[:, i, 1] = 1
phylowgs_cnv_info[:, i, 2] = 1
else:
if not np.all(c[:,l+g+i] == 1):
if np.max(c[:,l+g+i]) > 1:
print("2")
phylowgs_cnv_info[:, i, 0] = np.max(c[:,l+g+i])
phylowgs_cnv_info[:, i, 1] = np.round(np.mean(c[:, l+g+r+i][np.where(c[:, l+g+i] != 1)])).astype(int)
phylowgs_cnv_info[:, i, 2] = np.sum(u[:,np.where(c[:, l+g+i] != 1)[0]], axis=1)
elif np.min(c[:,l+g+i]) < 1:
print("3")
phylowgs_cnv_info[:, i, 0] = np.round(np.mean(c[:, l+g+r+i][np.where(c[:, l+g+i] != 1)])).astype(int)
phylowgs_cnv_info[:, i, 1] = np.min(c[:,l+g+i])
phylowgs_cnv_info[:, i, 2] = np.sum(u[:,np.where(c[:, l+g+i] != 1)[0]], axis=1)
elif not np.all(c[:,l+g+r+i] == 1):
if np.max(c[:,l+g+r+i]) > 1:
print("4")
phylowgs_cnv_info[:, i, 0] = np.max(c[:,l+g+r+i])
phylowgs_cnv_info[:, i, 1] = np.round(np.mean(c[:, l+g+i][np.where(c[:, l+g+r+i] != 1)])).astype(int)
phylowgs_cnv_info[:, i, 2] = np.sum(u[:,np.where(c[:, l+g+r+i] != 1)[0]], axis=1)
elif np.min(c[:,l+g+r+i]) < 1:
print("5")
phylowgs_cnv_info[:, i, 0] = np.round(np.mean(c[:, l+g+i][np.where(c[:, l+g+r+i] != 1)])).astype(int)
phylowgs_cnv_info[:, i, 1] = np.min(c[:,l+g+r+i])
phylowgs_cnv_info[:, i, 2] = np.sum(u[:,np.where(c[:, l+g+r+i] != 1)[0]], axis=1)
### generate F for SVs with randomness, given the generated CNVs
for (sv_chr, sv_pos, sv_idx) in sv_tuple_list:
cnv_idx = search_sv_cnv_num(sv_chr, sv_pos, seg_cn_idx_dict)
adj_cnv_idx = cnv_idx + int(int(d[sv_idx]) == int(bool_list[cnv_idx]))*r
adj_cnv_idx_a = cnv_idx + int(int(d[sv_idx]) != int(bool_list[cnv_idx])) * r
#print("p", F_true[:, sv_idx],F_true[:, l+g+adj_cnv_idx])
N_sv = np.random.binomial(N[:,adj_cnv_idx], F_true[:, sv_idx]/(F_true[:, l+g+adj_cnv_idx]))
ref_counts_sv = np.round(N[:,adj_cnv_idx] + N[:,adj_cnv_idx_a] - N_sv).astype(int)
A_sv[:, sv_idx] = N_sv
R_sv[:, sv_idx] = ref_counts_sv
for index in range(len(df_pyclone_list)):
df_pyclone_list[index] = df_pyclone_list[index].append({'mutation_id': 'sv' + str(sv_idx), 'ref_counts': ref_counts_sv[index], 'var_counts': N_sv[index], \
'normal_cn': 2, 'minor_cn': min(F_gen[index,l+g+adj_cnv_idx], F_gen[index,l+g+adj_cnv_idx_a]), \
'major_cn': max(F_gen[index,l+g+adj_cnv_idx], F_gen[index,l+g+adj_cnv_idx_a])}, ignore_index=True)
F_gen[:, sv_idx] = N_sv/read_depth
for i in range(0, len(snv_tuple_list)):
(snv_chr, snv_pos), snv_idx = snv_tuple_list[i]
if snv_idx in snv_idx_list:
snv_new_idx = np.where(snv_idx_list == snv_idx)[0][0]
#print(snv_new_idx, snv_pos,)
cnv_idx = search_sv_cnv_num(snv_chr, snv_pos, seg_cn_idx_dict)
adj_cnv_idx = cnv_idx + int(int(d[l+snv_new_idx]) == int(bool_list[cnv_idx])) * r
#print(c[:, l + g + adj_cnv_idx])
adj_cnv_idx_a = cnv_idx + int(int(d[l+snv_new_idx]) != int(bool_list[cnv_idx])) * r
#print("p", F_true[:, l + snv_new_idx], F_true[:, l + g + adj_cnv_idx], F_true[:, l + g + adj_cnv_idx_a] )
N_snv = np.random.binomial(N[:,adj_cnv_idx], F_true[:, l+snv_new_idx]/F_true[:, l+g+adj_cnv_idx])
ref_counts_snv = np.round((F_true[:, l + g + adj_cnv_idx] + F_true[:, l + g + adj_cnv_idx_a]) * N[:, cnv_idx] - N_snv).astype(int)
F_gen[:, l + snv_new_idx] = N_snv/read_depth
A_snv[:, snv_new_idx] = N_snv
R_snv[:, snv_new_idx] = ref_counts_snv
for index in range(len(df_pyclone_list)):
df_pyclone_list[index] = df_pyclone_list[index].append(
{'mutation_id': 'snv' + str(snv_idx), 'ref_counts': ref_counts_snv[index], 'var_counts': N_snv[index], \
'normal_cn': 2, 'minor_cn': min(F_gen[index, l + g + adj_cnv_idx], F_gen[index, l + g + adj_cnv_idx_a]), \
'major_cn': max(F_gen[index, l + g + adj_cnv_idx], F_gen[index, l + g + adj_cnv_idx_a])}, ignore_index=True)
#print(np.sqrt(np.mean(np.square(F_gen - F_true))))
#print(df_pyclone_list)
# reorder the allelic specific CNs of F and corresponding C
for i in range(0, r):
#print(F_true[:, l+g+i]/(F_true[:, l+g+i] + F_true[:, l+g+i+r]))
for j in range(F_gen.shape[0]):
if F_gen[j, l+g+i] < F_gen[j, l+g+r+i]:
temp = F_gen[j, l+g+i]
F_gen[j, l+g+i] = F_gen[j, l+g+r+i]
F_gen[j, l+g+r+i] = temp
return F_gen, df_pyclone_list, A_sv, A_snv, R_sv, R_snv, phylowgs_cnv_info
def generate_f_unsampled(u, c, c_cn, r, seg_cn_idx_dict, snv_cn_idx_dict, det, snv_un_idx_list, d_unsampled, bool_list, N, F_gen_cn):
print(u.shape, c.shape)
if det:
return np.dot(u, c)
else:
F_true_unsampled = np.dot(u, c)
F_gen_unsampled = np.zeros((F_true_unsampled.shape))
F_true_cn = np.dot(u, c_cn)
sv_tuple_list = []
snv_tuple_list = list(snv_cn_idx_dict.items())
n, _ = c.shape
### generate F for SVs with randomness
for i in range(0, len(snv_tuple_list)):
(snv_chr, snv_pos), snv_idx = snv_tuple_list[i]
if snv_idx in snv_un_idx_list:
snv_new_idx = np.where(snv_un_idx_list == snv_idx)[0][0]
cnv_idx = search_sv_cnv_num(snv_chr, snv_pos, seg_cn_idx_dict)
adj_cnv_idx = cnv_idx + int(int(d_unsampled[snv_new_idx]) == int(bool_list[cnv_idx])) * r
#print(F_true_unsampled[:, snv_new_idx], F_true_cn[:,adj_cnv_idx])
if F_true_unsampled[:, snv_new_idx] > F_true_cn[:,adj_cnv_idx]:
F_true_unsampled[:, snv_new_idx] - 1
F_gen_unsampled[:, snv_new_idx] = np.random.binomial((F_gen_cn[:, adj_cnv_idx] * N[:, cnv_idx]).astype(int),
F_true_unsampled[:, snv_new_idx] / F_true_cn[:,adj_cnv_idx]) / N[:,cnv_idx]
# F_gen[:, snv_new_idx] = np.random.binomial(N, F_true[:, snv_new_idx]/(F_cn[:, cnv_idx] + F_cn[:, cnv_idx+r]))/(N*(F_cn[:, cnv_idx] + F_cn[:, cnv_idx+r]))
return F_gen_unsampled
def search_sv_cnv_num(sv_chr, sv_pos, seg_cn_idx_dict):
for (bgn, end), cnv_idx in seg_cn_idx_dict[sv_chr].items():
if bgn <= sv_pos <= end:
return cnv_idx
# return matrix a, matrix h, and dictionary mate_dict
# matrix a: a ((2n-1) * l) matrix contains mated_reads info
# matrix h: a ((2n-1) * l) matrix contains total_reads info
# mate_dict:
# key: (chrom, pos, isLeft)
# val: (mate_chrom, mate_pos, mate_isLeft)
def get_a_h_mate_dict(tree, n, constants_dict):
l, sv_cn_idx_dict = get_bp_copy_num_idx_dict(tree, n, constants_dict)
a, h = make_2d_list(len(tree.node_list), l), make_2d_list(len(tree.node_list), l)
mate_dict = {}
for node_name in tree.node_list:
#print("node",node_name)
gene_prof = tree.idx_node_dict[node_name].geneProf
sv_dict = gene_prof.get_sv_read_nums_dict(constants_dict['cov'], constants_dict['read_len'])
for chrm in sv_dict.keys():
#print("ch",chrm)
for cur_pos, cur_is_left, cur_chr in sv_dict[chrm]:
mat_pos, mat_is_left, mat_chr = sv_dict[chrm][(cur_pos, cur_is_left, cur_chr)]['mate']
cur_key, mat_key = (cur_chr, cur_pos, cur_is_left), (mat_chr, mat_pos, mat_is_left)
#print(cur_key, mat_key)
if cur_key not in mate_dict:
mate_dict[cur_key] = mat_key
elif mate_dict[cur_key] != mat_key:
#print(sv_dict[cur_chr])
print 'There was an error generating SVs. Rerun sim.py until there are no errors.'
print 'cur_key:\t' + str(cur_key)
print 'mat_key:\t' + str(mat_key)
print 'cur_key was already mated with:\t' + str(mate_dict[cur_key])
exit()
if mat_key not in mate_dict:
mate_dict[mat_key] = cur_key
elif mate_dict[mat_key] != cur_key:
#print(sv_dict[mat_chr])
print 'There was an error generating SVs. Rerun sim.py until there are no errors.'
print 'cur_key:\t' + str(cur_key)
print 'mat_key:\t' + str(mat_key)
print 'mat_key was already mated with:\t' + str(mate_dict[mat_key])
exit()
j = sv_cn_idx_dict[chrm][(cur_pos, cur_is_left, cur_chr)]
a[node_name - 1][j] = sv_dict[chrm][(cur_pos, cur_is_left, cur_chr)]['mated_reads']
h[node_name - 1][j] = sv_dict[chrm][(cur_pos, cur_is_left, cur_chr)]['total_reads']
return a, h, mate_dict
# given a matrix, save as tsv file
def output_tsv(mtx, output_file, output_folder):
with open(output_folder + output_file, "w") as f:
f.write("\n".join("\t".join(map(str, x)) for x in mtx))
# given cnv_idx(int) and r(int), output rec_id(str)
# eg. given rec_idx = 1, r = 12, output 'cnv01'
def get_cnv_rec_id(cnv_idx, r):
return 'cnv' + str(cnv_idx).zfill(len(str(r)))
# given sv_idx(int) and r(int), output rec_id(str)
def get_sv_rec_id(sv_idx, l):
return 'sv' + str(sv_idx).zfill(len(str(l)))
### xf: add for snv
def get_snv_rec_id(snv_idx, g):
return 'snv' + str(snv_idx).zfill(len(str(g)))
# a, h, mate_dict = get_a_h_mate_dict(t, n, constants_dict)
# generate a vcf file for each sample
def generate_s_snv(metaFile, tree, l, sv_cn_idx_dict, r, seg_cn_idx_dict, g, snv_cn_idx_dict, snv_sampled_idx, snv_unsampled_idx,
seg_bgn_idx_dict, seg_end_idx_dict, F, F_unsampled_snv, U, C, c_p, c_m, a, h, mate_dict, outputFolder, sample_prob_idx,
A_sv, A_snv, R_sv, R_snv, phylowgs_cnv_info):
vcf_reader = vcf.Reader(open(metaFile, 'r'))
VAF_sv = A_sv / (A_sv + R_sv)
VAF_snv = A_snv / (A_snv + R_snv)
vcf_reader.metadata['filedate'][0] = datetime.datetime.now().date().strftime('%Y%m%d') # set date to current date
f_p = np.dot(U, c_p)
f_m = np.dot(U, c_m)
mixed_a = np.dot(U, a) # m * l
mixed_h = np.dot(U, h) # m * l
cntmd_dict = {}
df_phylowgs_list = []
for i in range(len(U)):
sample_idx = i + 1
cntmd_dict[sample_idx] = {}
df_phylowgs = pd.DataFrame(
columns=['chromosome', 'start', 'end', 'major_cn', 'minor_cn', 'cellular_prevalence'])
df_phylowgs_list.append(df_phylowgs)
if not os.path.exists(outputFolder + '/sample' + str(sample_prob_idx) + '/'):
os.mkdir(outputFolder + '/sample' + str(sample_prob_idx))
temp_file = outputFolder + '/sample' + str(sample_prob_idx) + '/sample' + str(sample_idx) + '.vcf'
temp_writer = vcf.Writer(open(temp_file, 'w'), vcf_reader)
temp_file_unsampled_snv = outputFolder + '/unsampled_snv_sample' + '_' + str(sample_prob_idx) + '_' + str(sample_idx) + '.vcf'
temp_writer_unsampled_snv = vcf.Writer(open(temp_file_unsampled_snv, 'w'), vcf_reader)
if not os.path.exists(outputFolder + '/phylowgs'):
os.mkdir(outputFolder + '/phylowgs')
temp_file_consensus_phylowgs = outputFolder + '/phylowgs/consensus_snv' + str(sample_idx) + '.vcf'
temp_writer_consensus_phylowgs = vcf.Writer(open(temp_file_consensus_phylowgs, 'w'), vcf_reader)
alt_type, gt_cnv = 'CNV', '1|1' # constants for all cnv records
for chrom in sorted(seg_cn_idx_dict.keys(),key=int):
cntmd_dict[sample_idx][chrom] = []
for (key, val) in sorted(seg_cn_idx_dict[chrom].items(), key = lambda x: int(x[1])):
pos = key[0]
rec_id = get_cnv_rec_id(val, r)
info_end = key[1]
cn = [F[i][l + len(snv_sampled_idx) + val], F[i][l + len(snv_sampled_idx) + r + val]]
cntmd_dict[sample_idx][chrom].append(F[i][l + len(snv_sampled_idx) + val] + F[i][l + len(snv_sampled_idx) + r + val])
temp_writer.write_record(generate_cnv(chrom, pos, rec_id, alt_type, info_end, gt_cnv, cn))
df_phylowgs_list[i] = df_phylowgs_list[i].append({'chromosome': chrom, 'start': pos,
'end': info_end, 'major_cn': int(phylowgs_cnv_info[i, val, 0]), 'minor_cn': int(phylowgs_cnv_info[i, val, 1]),
'cellular_prevalence': phylowgs_cnv_info[i, val, 2]}, ignore_index=True)
alt_ori, alt_cS, alt_wMA, gt_sv = True, str(), True, '1|0' # constants for all sv records
for chrom in sorted(sv_cn_idx_dict.keys(),key=int):
for (key, val) in sorted(sv_cn_idx_dict[chrom].items(), key = lambda x: x[1]):
pos, isLeft = key[0], key[1]
rec_id = get_sv_rec_id(val, l)
(mate_chrom, mate_pos, mate_isLeft) = mate_dict[(chrom, pos, isLeft)]
#print(sv_cn_idx_dict[mate_chrom])
mate_id = sv_cn_idx_dict[mate_chrom][(mate_pos, mate_isLeft, mate_chrom)]
alt_chr, alt_pos = mate_chrom, mate_pos
cnadj = F[i][val]
bdp, dp = int(round(mixed_a[i][val])), int(round(mixed_h[i][val]))
info_mateid = get_sv_rec_id(mate_id, l)
alt_rO = False if mate_isLeft == True else True
temp_writer.write_record(generate_sv(chrom, pos, rec_id, alt_chr, alt_pos, alt_ori, alt_rO, alt_cS, alt_wMA, info_mateid, gt_sv, cnadj, bdp, dp))
vaf = VAF_sv[i][val]
alt_counts_sv = A_snv[i][val]
ref_counts_sv = R_snv[i][val]
temp_writer_consensus_phylowgs.write_record(generate_snv_consensus(chrom, pos, rec_id,gt_sv, cnadj, vaf, alt_counts_sv, ref_counts_sv))
###xf: add snvs