-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathpdbstats.py
1004 lines (902 loc) · 39.1 KB
/
pdbstats.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
from proteinmath import *
from polypeptide import *
import os
import gc
import multiprocessing
from os.path import join
from functools import partial
from secondary_structure import *
from sparc_distribution import *
from charmm import *
from memory_profiler import profile
nonconsecutive_mode = 'n'
consecutive_mode = 'c'
hydrophobicity_mode = 'p'
alphacarbon_mode = 'a'
alphacarbon_consec_mode = 'ac'
default_network_mode = 'd' # Only supported by block_stats_network
sidechain_mode = 'sc' # Only supported by block_stats_network
secondary_structure_mode = 'ss' # Only supported by block_stats_network
permissible_sequences_mode = 'ps' # Only supported by block_stats_network
permissible_sequences_ss_mode = 'pss'
score_sequences_mode = 'scs' # Only supported by block_stats_network
both_orientations_mode = 'bo'
num_interacting_mode = 'ni'
random_coil_mode = 'rc'
def block_stats(id, mode, input, urls, output):
"""Calculates a block of PDB statistics based on the mode selected in the mode parameter. Modes are available above."""
if mode == nonconsecutive_mode or mode == consecutive_mode or mode == alphacarbon_mode or mode == alphacarbon_consec_mode:
data = [[[] for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)]
elif mode == hydrophobicity_mode:
data = [[] for i in range(AMINO_ACID_COUNT)]
peptide = Polypeptide()
for path in urls:
if path.find("pdb") == -1: continue
#print threading.currentThread().name, "processing", path, "..."
print "Processing " + path + " (%d)..." % id
try:
peptide.read(join(input, path))
except (AssertionError, ZeroDivisionError, ValueError):
print "==========================Omit %r" % path
continue
reported = False
try:
for i, aa in enumerate(peptide.aminoacids):
tag1 = aacode(aa.type)
if mode == nonconsecutive_mode or mode == alphacarbon_mode:
r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i, consec=False)
elif mode == consecutive_mode or mode == alphacarbon_consec_mode:
r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i, consec=True)
elif mode == hydrophobicity_mode:
r = peptide.nearby_aa(aa, 10.0, i)
if tag1 < 22:
data[tag1].append(len(r))
else:
print "Partial omit %r (unknown aa)." % path
reported = True
break
continue
for aa2 in r:
tag2 = aacode(aa2.type)
if tag1 < 22 and tag2 < 22:
if mode == alphacarbon_mode or mode == alphacarbon_consec_mode:
data[tag1][tag2].append(aa.tolocal(aa2.acarbon))
else:
data[tag1][tag2].append(aa.aa_position_zone(aa2))
elif not reported:
print "Partial omit %r (unknown aa)." % path
reported = True
break
except (AssertionError, ZeroDivisionError, ValueError):
if not reported:
print "Partial omit %r (exception)." % path
del peptide.aminoacids[:]
peptide.hashtable = None
gc.collect()
del peptide
if mode == nonconsecutive_mode or mode == consecutive_mode or mode == alphacarbon_mode or mode == alphacarbon_consec_mode:
if output is not None and not os.path.exists(output): os.mkdir(output)
for i in range(AMINO_ACID_COUNT):
for j in range(AMINO_ACID_COUNT):
if output is not None:
f = open(join(output, "%d-%d.txt" % (i, j)), 'w')
for datum in data[i][j]:
if output is not None:
f.write(str(datum).translate(None, "()") + "\n")
if output is not None:
f.close()
del data
elif mode == hydrophobicity_mode:
if output is not None and not os.path.exists(output): os.mkdir(output)
for i in range(AMINO_ACID_COUNT):
if output is not None:
f = open(join(output, "%d.txt" % i), 'w')
for datum in data[i]:
if output is not None:
f.write(str(datum) + '\n')
if output is not None:
f.close()
del data
def pool_block_stats(input, output, mode, (urls, i)=None):
block_stats(i, mode, input, urls, join(output, str(i)))
print "Done with batch", i
#12/8/14 8:44 pm: calculating the first 10 blocks (nullified)
#12/10/14 4:41 pm: calculated the first 3 blocks with improved memory management, starting blocks 3-22 inclusive
#12/10/14 9pm: it works! Now, blocks 23-end
#12/11/14 6:48 am: starting all blocks, consecutive
def pool_initializer():
pass #print "Starting", multiprocessing.current_process().name
def calculate_pdb_stats(input, output, mode=nonconsecutive_mode):
if output is not None and not os.path.exists(output): os.mkdir(output)
contents = os.listdir(input)
block_size = 100
i = 0
blocks = int(math.ceil(len(contents) / block_size))
print blocks, "blocks"
# Single thread program used to debug using memory_profiler
'''block_stats(0, input, [contents[1]], None)
block_stats(1, input, [contents[2]], None)
block_stats(2, input, [contents[3]], None)
return'''
pool = multiprocessing.Pool(processes=3, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_block_stats, input, output, mode)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, i + blocks)]
#print zipped
pool.map(partial_block_stats, zipped)
pool.close()
pool.join()
print "done"
def calculate_hydrophobicity_stats(input, output):
if not os.path.exists(output): os.mkdir(output)
contents = os.listdir(input)
block_size = 100
i = 0
blocks = int(math.ceil(len(contents) / block_size))
print blocks, "blocks"
# Single thread program used to debug using memory_profiler
'''block_stats(0, input, [contents[1]], None)
block_stats(1, input, [contents[2]], None)
block_stats(2, input, [contents[3]], None)
return'''
pool = multiprocessing.Pool(processes=3, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_block_stats, input, output, hydrophobicity_mode)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, i + blocks)]
#print zipped
pool.map(partial_block_stats, zipped)
pool.close()
pool.join()
print "done"
def calculate_pdb_alphacarbons(input, output, mode=alphacarbon_mode):
if not os.path.exists(output): os.mkdir(output)
contents = os.listdir(input)[:1000]
block_size = 100
i = 0
blocks = int(math.ceil(len(contents) / block_size))
print blocks, "blocks"
pool = multiprocessing.Pool(processes=3, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_block_stats, input, output, mode)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, i + blocks)]
#print zipped
pool.map(partial_block_stats, zipped)
pool.close()
pool.join()
print "done"
def calculate_amino_acid_counts(input):
contents = os.listdir(input)
data = [0 for i in xrange(AMINO_ACID_COUNT)]
peptide = Polypeptide()
for path in contents:
if path.find("pdb") == -1: continue
print "Processing %r..." % path
try:
peptide.read(join(input, path))
except (AssertionError, ZeroDivisionError, ValueError):
print "==========================Omit %r" % path
continue
for aa in peptide.aminoacids:
tag = aacode(aa.type)
if tag < 22: data[tag] += 1
del peptide.aminoacids[:]
peptide.hashtable = None
gc.collect()
del peptide
for x in data:
print x
print "Sum:", sum(data)
# MARK: Fall 2015, new code for a larger sample size
import requests, time, numpy, urllib2, sys
def block_stats_network(id, pdbids, output, mode='d', **kwargs):
"""Calculates a block of PDB statistics, and puts all the data into a directory specified by output. For mode, you may pass default_network_mode, sidechain_mode, or secondary_structure_mode."""
if mode == default_network_mode:
#Each pz will have an array of frequencies, where the frequencies correspond to the following situations:
#[distance1_default, helix1, helix5, helix7, sheet0, sheet1, sheet-1, distance2_default, helix1, ..., (up to distance5), nonconsec, helix1, ...]
data = {
#"default" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"long_range" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"short_range" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"consec" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"consec+secondary" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"secondary" : {
secondary_struct_helix + "1" : {}, secondary_struct_helix + "2" : {},
secondary_struct_helix + "3" : {}, secondary_struct_helix + "4" : {},
secondary_struct_helix + "5" : {}, secondary_struct_helix + "6" : {},
secondary_struct_helix + "7" : {}, secondary_struct_helix + "8" : {},
secondary_struct_helix + "9" : {}, secondary_struct_helix + "10" : {},
secondary_struct_sheet + "0" : {}, secondary_struct_sheet + "1" : {},
secondary_struct_sheet + "-1" : {}
},
"medium" : [[] for i in range(AMINO_ACID_COUNT)],
"permissible" : {
secondary_struct_helix + "1" : {}, secondary_struct_helix + "2" : {},
secondary_struct_helix + "3" : {}, secondary_struct_helix + "4" : {},
secondary_struct_helix + "5" : {}, secondary_struct_helix + "6" : {},
secondary_struct_helix + "7" : {}, secondary_struct_helix + "8" : {},
secondary_struct_helix + "9" : {}, secondary_struct_helix + "10" : {},
secondary_struct_sheet + "0" : {}, secondary_struct_sheet + "1" : {},
secondary_struct_sheet + "-1" : {}, "default": {}, "all": {}
}
}
elif mode == num_interacting_mode:
data = [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)]
elif mode == both_orientations_mode:
data = {
"long_range" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"short_range" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"consec" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"consec+secondary" : [[{} for i in range(AMINO_ACID_COUNT)] for j in range(AMINO_ACID_COUNT)],
"secondary" : {
secondary_struct_helix + "1" : {}, secondary_struct_helix + "2" : {},
secondary_struct_helix + "3" : {}, secondary_struct_helix + "4" : {},
secondary_struct_helix + "5" : {}, secondary_struct_helix + "6" : {},
secondary_struct_helix + "7" : {}, secondary_struct_helix + "8" : {},
secondary_struct_helix + "9" : {}, secondary_struct_helix + "10" : {},
secondary_struct_sheet + "0" : {}, secondary_struct_sheet + "1" : {},
secondary_struct_sheet + "-1" : {}
}
}
elif mode == sidechain_mode:
data = [{} for i in range(AMINO_ACID_COUNT)]
elif mode == secondary_structure_mode:
data = {
secondary_struct_helix + "1" : {}, secondary_struct_helix + "2" : {},
secondary_struct_helix + "3" : {}, secondary_struct_helix + "4" : {},
secondary_struct_helix + "5" : {}, secondary_struct_helix + "6" : {},
secondary_struct_helix + "7" : {}, secondary_struct_helix + "8" : {},
secondary_struct_helix + "9" : {}, secondary_struct_helix + "10" : {},
secondary_struct_sheet + "0" : {}, secondary_struct_sheet + "1" : {},
secondary_struct_sheet + "-1" : {}
}
elif mode == permissible_sequences_mode:
#It will contain tuples (zone1, zone2) for an amino acid pair as keys, and dictionaries of possible zones for the pair's neighbors as values.
data = {}
elif mode == permissible_sequences_ss_mode:
data = {
secondary_struct_helix + "1" : {}, secondary_struct_helix + "2" : {},
secondary_struct_helix + "3" : {}, secondary_struct_helix + "4" : {},
secondary_struct_helix + "5" : {}, secondary_struct_helix + "6" : {},
secondary_struct_helix + "7" : {}, secondary_struct_helix + "8" : {},
secondary_struct_helix + "9" : {}, secondary_struct_helix + "10" : {},
secondary_struct_sheet + "0" : {}, secondary_struct_sheet + "1" : {},
secondary_struct_sheet + "-1" : {}, "default": {}, "all": {}
}
elif mode == random_coil_mode:
data = {
secondary_struct_helix + "1" : {}, secondary_struct_helix + "2" : {},
secondary_struct_helix + "3" : {}, secondary_struct_helix + "4" : {},
secondary_struct_helix + "5" : {}, secondary_struct_helix + "6" : {},
secondary_struct_helix + "7" : {}, secondary_struct_helix + "8" : {},
secondary_struct_helix + "9" : {}, secondary_struct_helix + "10" : {},
secondary_struct_sheet + "0" : {}, secondary_struct_sheet + "1" : {},
secondary_struct_sheet + "-1" : {}, "default": {}, "all": {}
}
def add_to_data(aa, aa2, tag1, tag2, key, both=False, struct_type=None):
if tag1 < 22 and tag2 < 22:
array_mode = -1
if not key:
array_mode = int(min(math.fabs(aa2.tag - aa.tag), 6) - 1) * 7
key = "default"
pz = aa.aa_position_zone(aa2).alpha_zone
if both:
pz2 = aa2.aa_position_zone(aa).alpha_zone
if (pz, pz2) in data[key][tag1][tag2]:
data[key][tag1][tag2][(pz, pz2)] += 1
else:
data[key][tag1][tag2][(pz, pz2)] = 1
elif array_mode >= 0:
if struct_type:
struct_idx = next((i for i, ss in enumerate([None, "helix1", "helix5", "helix7", "sheet0", "sheet1", "sheet-1"]) if ss == struct_type), 0)
array_mode += struct_idx
if pz in data[key][tag1][tag2]:
data[key][tag1][tag2][pz][array_mode] += 1
else:
data[key][tag1][tag2][pz] = [0 for i in xrange(42)]
data[key][tag1][tag2][pz][array_mode] = 1
else:
if pz in data[key][tag1][tag2]:
data[key][tag1][tag2][pz] += 1
else:
data[key][tag1][tag2][pz] = 1
return True
else:
return False
peptide = Polypeptide()
for pdbid in pdbids:
pdbid = pdbid.strip()
print "Processing " + pdbid + " (%d)..." % id
try:
#response = urllib2.urlopen('http://www.rcsb.org/pdb/files/' + pdbid + '.pdb')
r = requests.get('http://www.rcsb.org/pdb/files/' + pdbid + '.pdb', headers={'user-agent': 'Mozilla/5.0'})
response = r.text.split('\n')
if len(response) <= 1:
assert False, "Got one-liner for PDB {}: {}".format(pdbid, r.text)
if mode == default_network_mode or mode == permissible_sequences_mode or mode == permissible_sequences_ss_mode or mode == both_orientations_mode:
peptide.read_file(response, secondary_structure=True, fillgaps=True)
elif mode == sidechain_mode:
peptide.read_file(response, otheratoms=True)
elif mode == secondary_structure_mode:
peptide.read_file(response, secondary_structure=True, fillgaps=True, fillends=True)
elif mode == random_coil_mode:
peptide.read_file(response, secondary_structure=True)
else:
peptide.read_file(response, secondary_structure=True, fillgaps=True)
del response
except Exception as e:
print "==========================Omit %r - %r" % (pdbid, e)
continue
#response.close()
#del response
reported = False
try:
if mode == secondary_structure_mode:
for sec_struct in peptide.secondary_structures:
for strand in sec_struct.strands:
prev_aa = peptide.aminoacids[strand.start]
for aa in peptide.aminoacids[strand.start + 1 : strand.end + 1]:
if not prev_aa or not aa:
prev_aa = aa
continue
pz = prev_aa.tolocal(aa.acarbon).floor()
pz2 = aa.tolocal(prev_aa.acarbon).floor()
if sec_struct.type + str(strand.identifiers[0]) not in data:
continue
if pz in data[sec_struct.type + str(strand.identifiers[0])]:
data[sec_struct.type + str(strand.identifiers[0])][pz] += 1
else:
data[sec_struct.type + str(strand.identifiers[0])][pz] = 1
if pz2 in data[sec_struct.type + str(strand.identifiers[0])]:
data[sec_struct.type + str(strand.identifiers[0])][pz2] += 1
else:
data[sec_struct.type + str(strand.identifiers[0])][pz2] = 1
prev_aa = aa
elif mode == num_interacting_mode:
#For each amino acid type, the dictionary will have the floored volume as its key, and a list of [observed interactions consec, obs short-range, obs long-range, possible interactions consec, possible short-range, possible long-range] as the value.
volume = math.floor(peptide.volume())
for aa in peptide.aminoacids:
if aa == None: continue
tag1 = aacode(aa.type)
#Find possible interactions as well as actually contacting residues
for aa2 in peptide.aminoacids:
if not aa2: continue
tag2 = aacode(aa2.type)
if tag1 > tag2 or (tag1 == tag2 and aa.tag > aa2.tag): continue
try:
if volume not in data[tag1][tag2]:
data[tag1][tag2][volume] = [0, 0, 0, 0, 0, 0]
idx = 0
if math.fabs(aa.tag - aa2.tag) == 1:
idx = 0
elif math.fabs(aa.tag - aa2.tag) <= 5:
idx = 1
else:
idx = 2
if aa.acarbon.distanceto(aa2.acarbon) <= 10.0:
data[tag1][tag2][volume][idx] += 1
data[tag1][tag2][volume][idx + 3] += 1
except:
print "Partial omit -", pdbid
elif mode == random_coil_mode:
iterations = 100
if "iterations" in kwargs: iterations = kwargs["iterations"]
assert "permissions" in kwargs, "No permissions object provided to block stats. Use the keyword argument 'permissions'."
permissions = kwargs["permissions"]
sec_struct_permissions = None
if "sec_struct_permissions" in kwargs: sec_struct_permissions = kwargs["sec_struct_permissions"]
for n in xrange(iterations):
#if n == iterations / 2: print n, "of", iterations, "iterations"
worked = peptide.randomcoil(permissions=permissions, struct_permissions=sec_struct_permissions, repeat_cutoff=10)
if not worked: continue
for aa in peptide.aminoacids:
r = peptide.nearby_aa(aa, 10.0, aa.tag)
for aa2 in r:
pz = aa.aa_position_zone(aa2).alpha_zone
separation = min(int(math.fabs(aa2.tag - aa.tag)), 6) - 1
sec_struct = peptide.secondary_structure_aa(aa.tag)
sec_struct_2 = peptide.secondary_structure_aa(aa2.tag)
if sec_struct and sec_struct_2 and sec_struct[1].start == sec_struct_2[1].start:
sec_name = sec_struct[0].type + str(sec_struct[1].identifiers[0])
if pz in data[sec_name]:
data[sec_name][pz][separation] += 1
else:
data[sec_name][pz] = [0 for i in xrange(1, 7)]
data[sec_name][pz][separation] = 1
else:
if pz in data["default"]:
data["default"][pz][separation] += 1
else:
data["default"][pz] = [0 for i in xrange(1, 7)]
data["default"][pz][separation] = 1
if pz in data["all"]:
data["all"][pz][separation] += 1
else:
data["all"][pz] = [0 for i in xrange(1, 7)]
data["all"][pz][separation] = 1
else:
for i, aa in enumerate(peptide.aminoacids):
if aa == None: continue
tag1 = aacode(aa.type)
if mode == permissible_sequences_mode or mode == permissible_sequences_ss_mode or mode == default_network_mode:
if i == 0 or i == len(peptide.aminoacids) - 1: continue
#Figure out the orientation pair between the last consecutive residue and this one
prev_aa = peptide.aminoacids[i - 1]
if not prev_aa: continue
zone1 = prev_aa.aa_position_zone(aa).alpha_zone
zone2 = aa.aa_position_zone(prev_aa).alpha_zone
sec_struct = peptide.secondary_structure_aa(i)
do_next = True
do_prev = True
if mode == default_network_mode:
cover_data = data["permissible"]
else:
cover_data = data
if mode == permissible_sequences_ss_mode or mode == default_network_mode:
if sec_struct:
if sec_struct[1].start > i - 1: continue
if sec_struct[1].start > i - 2: do_prev = False
if sec_struct[1].end < i + 1: do_next = False
sec_name = sec_struct[0].type + str(sec_struct[1].identifiers[0])
if sec_name not in cover_data: continue
data_lists = [cover_data[sec_name]]
else:
data_lists = [cover_data["default"]]
data_lists.append(cover_data["all"])
else:
data_list = [cover_data]
if peptide.aminoacids[i + 1] and do_next:
#Figure out the orientations between the next consecutive residue and this one
next_zone = aa.aa_position_zone(peptide.aminoacids[i + 1]).alpha_zone
for data_list in data_lists:
if (zone1, zone2) not in data_list:
data_list[(zone1, zone2)] = {}
if next_zone in data_list[(zone1, zone2)]:
data_list[(zone1, zone2)][next_zone] += 1
else:
data_list[(zone1, zone2)][next_zone] = 1
if peptide.aminoacids[i - 2] and do_prev:
#Figure out the orientations between the last two consecutive residues
prev_zone = prev_aa.aa_position_zone(peptide.aminoacids[i - 2]).alpha_zone
for data_list in data_lists:
if (zone1, zone2) not in data_list:
data_list[(zone1, zone2)] = {}
if prev_zone in data_list[(zone1, zone2)]:
data_list[(zone1, zone2)][prev_zone] += 1
else:
data_list[(zone1, zone2)][prev_zone] = 1
if mode == default_network_mode:
#Every orientation
'''r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i)
for aa2 in r:
tag2 = aacode(aa2.type)
sec_struct = peptide.secondary_structure_aa(aa.tag)
sec_struct_2 = peptide.secondary_structure_aa(aa2.tag)
if sec_struct and sec_struct_2 and sec_struct[1].start == sec_struct_2[1].start:
sec_name = sec_struct[0].type + str(sec_struct[1].identifiers[0])
else:
sec_name = None
if not add_to_data(aa, aa2, tag1, tag2, None, struct_type=sec_name):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break'''
#Nonconsec
r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i, consec=False)
for aa2 in r:
tag2 = aacode(aa2.type)
key = "long_range"
if math.fabs(aa2.tag - aa.tag) <= 5:
key = "short_range"
if not add_to_data(aa, aa2, tag1, tag2, key):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break
#Consec
r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i, consec=True)
for aa2 in r:
tag2 = aacode(aa2.type)
key = "consec"
sec_struct = peptide.secondary_structure_aa(i)
if sec_struct:
key = "secondary"
pz = aa.aa_position_zone(aa2).alpha_zone
sec_name = sec_struct[0].type + str(sec_struct[1].identifiers[0])
if sec_name in data[key]:
if pz in data[key][sec_name]:
data[key][sec_name][pz] += 1
else:
data[key][sec_name][pz] = 1
else:
if not add_to_data(aa, aa2, tag1, tag2, key):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break
if not add_to_data(aa, aa2, tag1, tag2, "consec+secondary"):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break
#Medium
r = peptide.nearby_aa(aa, 10.0, i)
if tag1 < 22:
data["medium"][tag1].append(len(r))
elif not reported:
print "Partial omit %r (unknown aa)." % pdbid
reported = True
if mode == both_orientations_mode:
#Nonconsec
r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i, consec=False)
for aa2 in r:
tag2 = aacode(aa2.type)
if tag1 > tag2 or (tag1 == tag2 and aa.tag > aa2.tag): continue
key = "long_range"
if math.fabs(aa2.tag - aa.tag) <= 5:
key = "short_range"
if not add_to_data(aa, aa2, tag1, tag2, key, both=True):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break
#Consec
r = peptide.nearby_aa(peptide.aminoacids[i], 10.0, i, consec=True)
for aa2 in r:
tag2 = aacode(aa2.type)
if tag1 > tag2 or (tag1 == tag2 and aa.tag > aa2.tag): continue
key = "consec"
sec_struct = peptide.secondary_structure_aa(i)
if sec_struct:
key = "secondary"
pz = aa.aa_position_zone(aa2).alpha_zone
pz2 = aa2.aa_position_zone(aa).alpha_zone
sec_name = sec_struct[0].type + str(sec_struct[1].identifiers[0])
if sec_name in data[key]:
if (pz, pz2) in data[key][sec_name]:
data[key][sec_name][(pz, pz2)] += 1
else:
data[key][sec_name][(pz, pz2)] = 1
else:
if not add_to_data(aa, aa2, tag1, tag2, key, both=True):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break
if not add_to_data(aa, aa2, tag1, tag2, "consec+secondary", both=True):
print "Partial omit %r (unknown aa)." % pdbid
reported = True
break
if mode == sidechain_mode:
if tag1 >= 22 and not reported:
print "Partial omit %r (unknown aa)." % pdbid
reported = True
continue
for atomname, location in aa.otheratoms.iteritems():
if atomname in data[tag1]:
data[tag1][atomname].append(location)
else:
data[tag1][atomname] = [location]
except (AssertionError, ZeroDivisionError, ValueError):
if not reported:
print "Partial omit %r (exception)." % pdbid
del peptide.aminoacids[:]
peptide.hashtable.clear()
gc.collect()
del peptide
if output is None: return
if not os.path.exists(output): os.mkdir(output)
def write_data(key, twopzs=False):
write_path = join(output, key)
if not os.path.exists(write_path): os.mkdir(write_path)
for i in range(AMINO_ACID_COUNT):
for j in range(AMINO_ACID_COUNT):
f = open(join(write_path, "%d-%d.txt" % (i, j)), 'w')
if twopzs:
for pzs, freq in data[key][i][j].iteritems():
f.write(str(pzs[0].x) + ", " + str(pzs[0].y) + ", " + str(pzs[0].z) + ", " + str(pzs[1].x) + ", " + str(pzs[1].y) + ", " + str(pzs[1].z) + "; " + str(freq) + "\n")
else:
for pz, freq in data[key][i][j].iteritems():
if isinstance(freq, int):
freqstr = str(freq)
else:
freqstr = ""
for fr in freq:
freqstr += str(fr) + ","
freqstr = freqstr[:-1]
f.write(str(pz.x) + ", " + str(pz.y) + ", " + str(pz.z) + "; " + freqstr + "\n")
f.close()
if mode == default_network_mode:
write_data("long_range")
write_data("consec")
write_data("short_range")
write_data("consec+secondary")
#Secondary
write_path = join(output, "secondary")
if not os.path.exists(write_path): os.mkdir(write_path)
for sec_struct_type in data["secondary"]:
f = open(join(write_path, sec_struct_type + ".txt"), 'w')
for pz, freq in data["secondary"][sec_struct_type].iteritems():
f.write(str(pz.x) + ", " + str(pz.y) + ", " + str(pz.z) + "; " + str(freq) + "\n")
f.close()
#write_data("default")
#Medium
medium_path = join(output, "medium")
if not os.path.exists(medium_path): os.mkdir(medium_path)
for i in range(AMINO_ACID_COUNT):
f = open(join(medium_path, "%d.txt" % i), 'w')
for datum in data["medium"][i]:
f.write(str(datum) + '\n')
f.close()
#Permissible sequences
permissibles_path = os.path.join(output, "permissible_sequences")
if not os.path.exists(permissibles_path): os.mkdir(permissibles_path)
for sec_struct_type in data["permissible"]:
with open(os.path.join(permissibles_path, sec_struct_type + ".txt"), "w") as file:
for zone1, zone2 in data["permissible"][sec_struct_type]:
file.write("{}, {}, {}; {}, {}, {}\n".format(zone1.x, zone1.y, zone1.z, zone2.x, zone2.y, zone2.z))
for pz, freq in data["permissible"][sec_struct_type][(zone1, zone2)].iteritems():
file.write(str(pz.x) + ", " + str(pz.y) + ", " + str(pz.z) + "; " + str(freq) + "\n")
file.write("\n")
elif mode == random_coil_mode:
for sec_struct_type in data:
f = open(join(output, sec_struct_type + ".txt"), "w")
for pz, freqs in data[sec_struct_type].iteritems():
freqstr = ""
for freq in freqs: freqstr += str(freq) + ","
freqstr = freqstr[:-1]
f.write("{:.0f},{:.0f},{:.0f};{}\n".format(pz.x, pz.y, pz.z, freqstr))
f.close()
elif mode == both_orientations_mode:
write_data("long_range", twopzs=True)
write_data("consec", twopzs=True)
write_data("short_range", twopzs=True)
write_data("consec+secondary", twopzs=True)
#Secondary
write_path = join(output, "secondary")
if not os.path.exists(write_path): os.mkdir(write_path)
for sec_struct_type in data["secondary"]:
f = open(join(write_path, sec_struct_type + ".txt"), 'w')
for pzs, freq in data["secondary"][sec_struct_type].iteritems():
f.write(str(pzs[0].x) + ", " + str(pzs[0].y) + ", " + str(pzs[0].z) + ", " + str(pzs[1].x) + ", " + str(pzs[1].y) + ", " + str(pzs[1].z) + "; " + str(freq) + "\n")
f.close()
elif mode == num_interacting_mode:
for i in range(AMINO_ACID_COUNT):
for j in range(AMINO_ACID_COUNT):
f = open(join(output, "%d-%d.txt" % (i, j)), 'w')
for volume, counts in data[i][j].iteritems():
f.write("{};{},{},{},{},{},{}\n".format(volume, counts[0], counts[1], counts[2], counts[3], counts[4], counts[5]))
f.close()
elif mode == permissible_sequences_mode:
write_path = join(output, "permissible_sequences.txt")
with open(write_path, "w") as file:
for zone1, zone2 in data:
file.write("{}, {}, {}; {}, {}, {}\n".format(zone1.x, zone1.y, zone1.z, zone2.x, zone2.y, zone2.z))
for pz, freq in data[(zone1, zone2)].iteritems():
file.write(str(pz.x) + ", " + str(pz.y) + ", " + str(pz.z) + "; " + str(freq) + "\n")
file.write("\n")
elif mode == permissible_sequences_ss_mode:
for sec_struct_type in data:
with open(os.path.join(output, sec_struct_type + ".txt"), "w") as file:
for zone1, zone2 in data[sec_struct_type]:
file.write("{}, {}, {}; {}, {}, {}\n".format(zone1.x, zone1.y, zone1.z, zone2.x, zone2.y, zone2.z))
for pz, freq in data[sec_struct_type][(zone1, zone2)].iteritems():
file.write(str(pz.x) + ", " + str(pz.y) + ", " + str(pz.z) + "; " + str(freq) + "\n")
file.write("\n")
elif mode == sidechain_mode:
for i in range(AMINO_ACID_COUNT):
f = open(join(output, "%d.txt" % i), 'w')
for atomname, locations in data[i].iteritems():
totalpoint = Point3D.zero()
for location in locations:
totalpoint = totalpoint.add(location)
totalpoint = totalpoint.multiply(1.0 / len(locations))
distance = 0.0
for location in locations:
distance += totalpoint.distanceto(location)
distance /= float(len(locations))
f.write(atomname + ',' + str(totalpoint.x) + ',' + str(totalpoint.y) + ',' + str(totalpoint.z) + ',' + str(distance) + '\n')
f.close()
elif mode == secondary_structure_mode:
for struct_type in data:
out_path = join(output, struct_type + ".txt")
with open(out_path, "w") as f:
for pz, freq in data[struct_type].iteritems():
f.write(str(pz.x) + ", " + str(pz.y) + ", " + str(pz.z) + "; " + str(freq) + "\n")
def pool_block_stats_network(input, output, mode, (urls, i)=None):
if output:
if os.path.exists(join(output, str(i))): return
block_stats_network(i, urls, join(output, str(i)), mode=mode)
else:
block_stats_network(i, urls, None, mode=mode)
print "Done with batch", i
gc.collect()
time.sleep(60)
def calculate_pdb_stats_network(input, output, mode=default_network_mode):
"""Calculates pdb stats by retrieving files with the PDB IDs listed in the newline-separated input file."""
if output is not None and not os.path.exists(output): os.mkdir(output)
with open(input, 'r') as file:
contents = file.readlines()
#contents = numpy.random.choice(contents, 500)
block_size = 50
i = 0
blocks = int(math.ceil(len(contents) / block_size))
if mode == default_network_mode:
rate = 1.8
time = rate * len(contents)
time_unit = "seconds"
if time > 60:
time /= 60.0
time_unit = "minutes"
if time > 60:
time /= 60.0
time_unit = "hours"
print "NOTE: The calculation process will be performed in {}-structure blocks. Generally this takes roughly {} seconds per structure, so with {} blocks your job will likely take about {:.2f} {}.".format(block_size, rate, blocks, time, time_unit)
pool = multiprocessing.Pool(processes=3, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_block_stats_network, input, output, mode)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, blocks)]
#print zipped
pool.map_async(partial_block_stats, zipped)
pool.close()
pool.join()
print "done"
def num_possible_interacting(pdbsample, output):
"""Writes the number of possible interactions each amino acid pair could have in the PDB sample."""
calculate_pdb_stats_network(pdbsample, output, mode=num_interacting_mode)
#MARK: Structure pairs
def generate_structurepairs(pdbsample, output):
"""Generates PDB files according with each pair of amino acid types from the pdb sample. These PDB files contain all sidechain atoms from that structure."""
if not os.path.exists(output):
os.mkdir(output)
with open(pdbsample, 'r') as file:
contents = file.readlines()
allcombos = [(i, j) for l in [[(i, j) for j in xrange(i, AMINO_ACID_COUNT)] for i in xrange(AMINO_ACID_COUNT)] for i, j in l if i != 3 and i != 8 and j != 3 and j != 8]
for pdbid in contents:
pdbid = pdbid.strip()
print "Processing " + pdbid, ",", len(allcombos), "combinations left..."
try:
response = urllib2.urlopen('http://www.rcsb.org/pdb/files/' + pdbid + '.pdb')
peptide = Polypeptide()
peptide.read_file(response, otheratoms=True)
response.close()
except:
print "==========================Omit %r" % pdbid
continue
del response
for i, aa in enumerate(peptide.aminoacids):
for aa2 in peptide.aminoacids[i + 1:]:
type1 = aacode(aa.type)
type2 = aacode(aa2.type)
if type1 < type2: combo = (type1, type2)
else: combo = (type2, type1)
if combo in allcombos:
print combo
newpeptide = Polypeptide(preaas=[aa, aa2])
newpeptide.center()
with open(os.path.join(output, str(type1) + "-" + str(type2) + ".pdb"), "w") as file:
file.write(newpeptide.pdb())
allcombos.remove(combo)
del newpeptide
del peptide
gc.collect()
if len(allcombos) == 0:
break
print "Done"
def sparc_score_sequences(pdb_path, output, distributions, stop_pt=None, charmm_scores=False):
tmpdir = os.path.join(os.path.dirname(pdb_path), "sp_tmp")
if charmm_scores:
if os.path.exists(tmpdir):
shutil.rmtree(tmpdir)
os.mkdir(tmpdir)
peptide = Polypeptide()
pdbids = []
with open(pdb_path, "r") as file:
found_stopping_pt = (stop_pt == None or len(stop_pt) == 0)
for line in file:
if len(line.strip()) == 0: continue
if not found_stopping_pt and line.strip() == stop_pt:
found_stopping_pt = True
continue
if not found_stopping_pt: continue
pdbids.append(line.strip())
for pdbid in pdbids:
pdbid = pdbid.strip()
if random.randint(0, 100) < 80: continue
print "Processing " + pdbid + "..."
try:
response = urllib2.urlopen('http://www.rcsb.org/pdb/files/' + pdbid + '.pdb')
peptide.read_file(response, secondary_structure=True, fillgaps=True, otheratoms=charmm_scores)
except:
print "==========================Omit %r" % pdbid
continue
response.close()
del response
for i in xrange(len(peptide.aminoacids)):
if random.randint(0, 100) < 90: continue
length = random.randint(3, 10)
length = min(len(peptide.aminoacids), i + length) - i
try:
print "Evaluating segment", i, "-", i + length
scores = [d.score(peptide, peptide.aminoacids[i:i + length], isolate=True) for d in distributions]
if charmm_scores == True:
npeptide = Polypeptide()
npeptide.add_aas(peptide.aminoacids[i:i + length])
cscores = charmm_score(npeptide, os.path.join(tmpdir, "sequence.pdb"), mdpfile="../../default_nonminimized.mdp")
for c in cscores:
scores.append(c)
with open(output, "a") as file:
file.write("{},{},".format(pdbid, length))
scorestr = ""
for s in scores: scorestr += str(s) + ","
file.write(scorestr[:-1] + "\n")
except:
print "Partial omit -", pdbid
del peptide.aminoacids[:]
peptide.hashtable = None
gc.collect()
del peptide
if charmm_scores:
shutil.rmtree(tmpdir)
def pool_pdb_reference_state(input, output, mode, iterations, permissions, sec_struct_permissions, (urls, i)=None):
if output:
if os.path.exists(join(output, str(i))): return
block_stats_network(i, urls, join(output, str(i)), mode=mode, iterations=iterations, permissions=permissions, sec_struct_permissions=sec_struct_permissions)
else:
block_stats_network(i, urls, None, mode=mode, iterations=iterations, permissions=permissions, sec_struct_permissions=sec_struct_permissions)
print "Done with batch", i
gc.collect()
time.sleep(60)
def pdb_reference_state(input, output, permissions, sec_struct_permissions, iterations=1):
"""Calculates the SPARC reference state by retrieving files with the PDB IDs listed in the newline-separated input file. Each structure undergoes a default of 20 random coils, and the orientations found in each structure are recorded."""
if output is not None and not os.path.exists(output): os.mkdir(output)
with open(input, 'r') as file:
contents = file.readlines()
#contents = numpy.random.choice(contents, 500)
block_size = 10
i = 0
blocks = int(math.ceil(len(contents) / block_size))
pool = multiprocessing.Pool(processes=4, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_pdb_reference_state, input, output, random_coil_mode, iterations, permissions, sec_struct_permissions)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, blocks)]
#print zipped
pool.map_async(partial_block_stats, zipped)
pool.close()
pool.join()
print "done"
#MARK: Secondary Structure Analysis
def analyze_secondary_structure(input, output, sample=-1):
"""Retrieves files with the PDB IDs listed in the newline-separated input file, then determines the relative orientations going forward through each secondary structure in the A chain. The output is a file for each secondary structure type (10 types of helices, and initial, parallel, and anti-parallel sheets)."""
if output is not None and not os.path.exists(output): os.mkdir(output)
with open(input, 'r') as file:
contents = file.readlines()
if sample > 0:
contents = random.sample(contents, sample)
block_size = 50
i = 0
blocks = int(math.ceil(len(contents) / block_size))
print blocks, "blocks"
pool = multiprocessing.Pool(processes=3, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_block_stats_network, input, output, secondary_structure_mode)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, blocks)]
pool.map(partial_block_stats, zipped)
pool.close()
pool.join()
print "done"
def generate_permissible_sequences(input, output, sample=-1):
"""Retrieves files with the PDB IDs listed in the newline-separated input file, then determines the sets of permissible orientations between triplets of amino acids. The format of the return files will be
zone1; zone2
pz; freq
pz; freq
pz; freq
.
.
.
zone1; zone2
pz; freq
...
For example, given amino acids ABCD, zone1 and zone2 refer to B and C's orientation w.r.t. each other. pz could refer to the orientation between C and D or B and A."""
if output is not None and not os.path.exists(output): os.mkdir(output)
with open(input, 'r') as file:
contents = file.readlines()
if sample > 0:
contents = random.sample(contents, sample)
block_size = 50
i = 0
blocks = int(math.ceil(len(contents) / float(block_size)))
pool = multiprocessing.Pool(processes=3, initializer=pool_initializer, maxtasksperchild=1)
partial_block_stats = partial(pool_block_stats_network, input, output, permissible_sequences_ss_mode)
zipped = [(contents[block_size * k : min(block_size * (k + 1), len(contents))], k) for k in xrange(i, blocks)]
pool.map(partial_block_stats, zipped)