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
|
// $Id$
//
// Copyright (C) 2002-2012 Greg Landrum and Rational Discovery LLC
//
// @@ All Rights Reserved @@
// This file is part of the RDKit.
// The contents are covered by the terms of the BSD license
// which is included in the file license.txt, found at the root
// of the RDKit source tree.
//
#include <GraphMol/RDKitBase.h>
#include <RDGeneral/utils.h>
#include <GraphMol/RankAtoms.h>
#include <boost/cstdint.hpp>
#include <boost/foreach.hpp>
#include <boost/dynamic_bitset.hpp>
#include <RDGeneral/hash/hash.hpp>
#include <boost/lambda/lambda.hpp>
#include <list>
#include <algorithm>
using namespace boost::lambda;
//#define VERBOSE_CANON 1
//#define VERYVERBOSE_CANON 1
namespace RankAtoms{
using namespace RDKit;
// --------------------------------------------------
//
// grabs the corresponding primes for the rank vector ranks
//
// --------------------------------------------------
void getPrimes(const INT_VECT &ranks,INT_VECT &res){
PRECONDITION(res.size()==0,"");
res.reserve(ranks.size());
for(INT_VECT_CI ivCIt=ranks.begin();ivCIt!=ranks.end();++ivCIt){
res.push_back(firstThousandPrimes[(*ivCIt)%NUM_PRIMES_AVAIL]);
}
}
// --------------------------------------------------
//
// blows out any indices in indicesInPlay which correspond to unique ranks
//
// --------------------------------------------------
void updateInPlayIndices(const INT_VECT &ranks,INT_LIST &indicesInPlay){
INT_LIST::iterator ivIt=indicesInPlay.begin();
while(ivIt!=indicesInPlay.end()){
// find the first instance of this rank:
INT_VECT::const_iterator pos=std::find(ranks.begin(),ranks.end(),ranks[*ivIt]);
++pos;
// now check to see if there is at least one more:
if( std::find(pos,ranks.end(),ranks[*ivIt])==ranks.end()){
INT_LIST::iterator tmpIt = ivIt;
++ivIt;
indicesInPlay.erase(tmpIt);
} else {
++ivIt;
}
}
}
// --------------------------------------------------
//
// for each index in indicesInPlay, generate the products of the adjacent
// elements
//
// The products are weighted by the order of the bond connecting the atoms.
//
// --------------------------------------------------
void calcAdjacentProducts(unsigned int nAtoms,
const INT_VECT &valVect,
double const *adjMat,
const INT_LIST &indicesInPlay,
DOUBLE_VECT &res,
bool useSelf=true,
double tol=1e-6){
PRECONDITION(valVect.size() >= nAtoms,"");
PRECONDITION(res.size() == 0,"");
PRECONDITION(adjMat,"");
for(INT_LIST::const_iterator idxIt=indicesInPlay.begin();
idxIt != indicesInPlay.end();
++idxIt){
double accum;
if(useSelf)
accum=valVect[*idxIt];
else
accum=1.0;
const unsigned int iTab = (*idxIt)*nAtoms;
for(unsigned int j=0;j<nAtoms;++j){
double elem=adjMat[iTab+j];
if(elem>tol){
if(elem<2.-tol){
accum *= valVect[j];
} else {
accum *= pow(static_cast<double>(valVect[j]),
static_cast<int>(elem));
}
}
}
res.push_back(accum);
}
}
template <typename T>
void debugVect(const std::vector<T> arg){
typename std::vector<T>::const_iterator viIt;
for(viIt=arg.begin();viIt!=arg.end();++viIt){
BOOST_LOG(rdDebugLog)<< *viIt << " ";
}
BOOST_LOG(rdDebugLog)<< std::endl;
}
// --------------------------------------------------
//
// This is one round of the process from Step III in the Daylight
// paper
//
// --------------------------------------------------
unsigned int iterateRanks(unsigned int nAtoms,INT_VECT &primeVect,
DOUBLE_VECT &atomicVect,
INT_LIST &indicesInPlay,
double *adjMat,
INT_VECT &ranks,
VECT_INT_VECT *rankHistory,unsigned int stagnantTol){
PRECONDITION(!rankHistory||rankHistory->size()>=nAtoms,"bad rankHistory size");
bool done = false;
unsigned int numClasses = countClasses(ranks);
unsigned int lastNumClasses = 0;
unsigned int nCycles = 0;
unsigned int nStagnant=0;
//
// loop until either we finish or no improvement is seen
//
#ifdef VERBOSE_CANON
for(unsigned int i=0;i<nAtoms;i++) {
BOOST_LOG(rdDebugLog)<< "\t\t>:" << i << " " << ranks[i] << std::endl;
}
BOOST_LOG(rdDebugLog)<< "\t\t-*-*-*-*-" << std::endl;
#endif
while(!done && nCycles < nAtoms){
// determine which atomic indices are in play (which have duplicate ranks)
if(rankHistory){
for(INT_LIST_CI idx=indicesInPlay.begin();idx!=indicesInPlay.end();++idx){
(*rankHistory)[*idx].push_back(ranks[*idx]);
}
}
updateInPlayIndices(ranks,indicesInPlay);
if(indicesInPlay.empty()) break;
#ifdef VERYVERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "IN PLAY:" << std::endl;
BOOST_LOG(rdDebugLog)<< "\t\t->";
for(INT_LIST::const_iterator tmpI=indicesInPlay.begin();tmpI != indicesInPlay.end();tmpI++){
BOOST_LOG(rdDebugLog)<< " " << *tmpI;
}
BOOST_LOG(rdDebugLog)<< std::endl;
BOOST_LOG(rdDebugLog)<< "\t\t---------" << std::endl;
#endif
//-------------------------
// Step (2):
// Get the products of adjacent primes
//-------------------------
primeVect.resize(0);
getPrimes(ranks,primeVect);
atomicVect.resize(0);
calcAdjacentProducts(nAtoms,primeVect,adjMat,indicesInPlay,atomicVect,false);
#ifdef VERYVERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "primes: ";
debugVect(primeVect);
BOOST_LOG(rdDebugLog)<< "products: ";
debugVect(atomicVect);
#endif
//-------------------------
// Steps (3) and (4)
// sort the products and count classes
//-------------------------
sortAndRankVect(nAtoms,atomicVect,indicesInPlay,ranks);
lastNumClasses = numClasses;
numClasses = countClasses(ranks);
if(numClasses == lastNumClasses) nStagnant++;
#ifdef VERYVERBOSE_CANON
int tmpOff=0;
for(unsigned int i=0;i<nAtoms;i++){
//for(INT_LIST::const_iterator tmpI=indicesInPlay.begin();tmpI != indicesInPlay.end();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\ti:" << i << "\t" << ranks[i] << "\t" << primeVect[i];
if(std::find(indicesInPlay.begin(),indicesInPlay.end(),i)!=indicesInPlay.end()){
BOOST_LOG(rdDebugLog)<< "\t" << atomicVect[tmpOff];
tmpOff++;
}
BOOST_LOG(rdDebugLog)<< std::endl; }
BOOST_LOG(rdDebugLog)<< "\t\t---------" << std::endl;
#endif
// terminal condition, we'll allow a single round of stagnancy
if(numClasses == nAtoms || nStagnant > stagnantTol) done = 1;
nCycles++;
}
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< ">>>>>> done inner iteration. static: "<< nStagnant << " ";
BOOST_LOG(rdDebugLog)<< nCycles << " " << nAtoms << " " << numClasses << std::endl;
#ifdef VERYVERBOSE_CANON
for(unsigned int i=0;i<nAtoms;i++) {
BOOST_LOG(rdDebugLog)<< "\t" << i << " " << ranks[i] << std::endl;
}
#endif
if(nCycles == nAtoms){
BOOST_LOG(rdWarningLog) << "WARNING: ranking bottomed out" << std::endl;
}
#endif
return numClasses;
}
unsigned int iterateRanks2(unsigned int nAtoms,INT_VECT &primeVect,
DOUBLE_VECT &atomicVect,
INT_LIST &indicesInPlay,
double *adjMat,
INT_VECT &ranks,VECT_DOUBLE_VECT &nRanks,
VECT_INT_VECT *rankHistory,unsigned int stagnantTol){
PRECONDITION(!rankHistory||rankHistory->size()>=nAtoms,"bad rankHistory size");
bool done = false;
unsigned int numClasses = countClasses(ranks);
unsigned int lastNumClasses = 0;
unsigned int nCycles = 0;
unsigned int nStagnant=0;
//
// loop until either we finish or no improvement is seen
//
#ifdef VERBOSE_CANON
for(unsigned int i=0;i<nAtoms;i++) {
BOOST_LOG(rdDebugLog)<< "\t\t>:" << i << " " << ranks[i] << std::endl;
}
BOOST_LOG(rdDebugLog)<< "\t\t-*-*-*-*-" << std::endl;
#endif
while(!done && nCycles < nAtoms){
// determine which atomic indices are in play (which have duplicate ranks)
if(rankHistory){
BOOST_FOREACH(int idx,indicesInPlay){
(*rankHistory)[idx].push_back(ranks[idx]);
}
}
updateInPlayIndices(ranks,indicesInPlay);
if(indicesInPlay.empty()) break;
#ifdef VERYVERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "IN PLAY:" << std::endl;
BOOST_LOG(rdDebugLog)<< "\t\t->";
for(INT_LIST::const_iterator tmpI=indicesInPlay.begin();tmpI != indicesInPlay.end();tmpI++){
BOOST_LOG(rdDebugLog)<< " " << *tmpI;
}
BOOST_LOG(rdDebugLog)<< std::endl;
BOOST_LOG(rdDebugLog)<< "\t\t---------" << std::endl;
#endif
//-------------------------
// Step (2):
// Get the products of adjacent primes
//-------------------------
primeVect.resize(0);
getPrimes(ranks,primeVect);
atomicVect.resize(0);
calcAdjacentProducts(nAtoms,primeVect,adjMat,indicesInPlay,atomicVect,false);
#ifdef VERYVERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "primes: ";
debugVect(primeVect);
BOOST_LOG(rdDebugLog)<< "products: ";
debugVect(atomicVect);
#endif
unsigned int p=0;
BOOST_FOREACH(int idx,indicesInPlay){
nRanks[idx].push_back(atomicVect[p++]);
}
#ifdef VERYVERBOSE_CANON
for(int idx=0;idx<nAtoms;++idx){
std::cerr<<" nranks["<<idx<<"]: ";
std::copy(nRanks[idx].begin(),nRanks[idx].end(),std::ostream_iterator<double>(std::cerr," "));
std::cerr<<"\n";
}
#endif
//-------------------------
// Steps (3) and (4)
// sort the products and count classes
//-------------------------
rankVect(nRanks,ranks);
//sortAndRankVect2(nRanks,indicesInPlay,ranks);
lastNumClasses = numClasses;
numClasses = countClasses(ranks);
if(numClasses == lastNumClasses) nStagnant++;
#ifdef VERYVERBOSE_CANON
int tmpOff=0;
for(unsigned int i=0;i<nAtoms;i++){
//for(INT_LIST::const_iterator tmpI=indicesInPlay.begin();tmpI != indicesInPlay.end();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\ti:" << i << "\t" << ranks[i] << "\t" << primeVect[i];
if(std::find(indicesInPlay.begin(),indicesInPlay.end(),i)!=indicesInPlay.end()){
BOOST_LOG(rdDebugLog)<< "\t" << atomicVect[tmpOff];
tmpOff++;
}
BOOST_LOG(rdDebugLog)<< std::endl; }
BOOST_LOG(rdDebugLog)<< "\t\t---------" << std::endl;
#endif
// terminal condition, we'll allow a single round of stagnancy
if(numClasses == nAtoms || nStagnant > stagnantTol) done = 1;
nCycles++;
}
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< ">>>>>> done inner iteration. static: "<< nStagnant << " ";
BOOST_LOG(rdDebugLog)<< nCycles << " " << nAtoms << " " << numClasses << std::endl;
#ifdef VERYVERBOSE_CANON
for(unsigned int i=0;i<nAtoms;i++) {
BOOST_LOG(rdDebugLog)<< "\t" << i << " " << ranks[i] << std::endl;
}
#endif
if(nCycles == nAtoms){
BOOST_LOG(rdWarningLog) << "WARNING: ranking bottomed out" << std::endl;
}
#endif
return numClasses;
}
// --------------------------------------------------
//
// Calculates invariants for the atoms of a molecule
//
// NOTE: if the atom has not had chirality info pre-calculated, it doesn't
// much matter what value includeChirality has!
// --------------------------------------------------
void buildAtomInvariants(const ROMol &mol,INVAR_VECT &res,
bool includeChirality,
bool includeIsotopes){
PRECONDITION(res.size()>=mol.getNumAtoms(),"res vect too small");
unsigned int atsSoFar=0;
std::vector<boost::uint64_t> tres(mol.getNumAtoms());
for(ROMol::ConstAtomIterator atIt=mol.beginAtoms();atIt!=mol.endAtoms();atIt++){
Atom const *atom = *atIt;
int nHs = atom->getTotalNumHs() % 8;
int chg = abs(atom->getFormalCharge()) % 8;
int chgSign = atom->getFormalCharge() > 0;
int num = atom->getAtomicNum() % 128;
int nConns = atom->getDegree() % 8;
int deltaMass=0;
if(includeIsotopes && atom->getIsotope()){
deltaMass = static_cast<int>(atom->getIsotope() -
PeriodicTable::getTable()->getMostCommonIsotope(atom->getAtomicNum()));
deltaMass += 128;
if(deltaMass < 0) deltaMass = 0;
else deltaMass = deltaMass % 256;
}
// figure out the minimum-sized ring we're involved in
int inRing = 0;
if(atom->getOwningMol().getRingInfo()->numAtomRings(atom->getIdx())){
RingInfo *ringInfo=atom->getOwningMol().getRingInfo();
inRing=3;
while(inRing<256){
if(ringInfo->isAtomInRingOfSize(atom->getIdx(),inRing)){
break;
} else {
inRing++;
}
}
}
inRing = inRing % 16;
boost::uint64_t invariant = 0;
invariant = (invariant << 3) | nConns;
// we used to include the number of explicitHs, but that
// didn't make much sense. TotalValence is another possible
// discriminator here, but the information is essentially
// redundant with nCons, num, and nHs.
// invariant = (invariant << 4) | totalVal;
invariant = (invariant << 7) | num;
invariant = (invariant << 8) | deltaMass;
invariant = (invariant << 3) | nHs;
invariant = (invariant << 4) | inRing;
invariant = (invariant << 3) | chg;
invariant = (invariant << 1) | chgSign;
if(includeChirality ){
int isR=0;
if( atom->hasProp("_CIPCode")){
std::string cipCode;
atom->getProp("_CIPCode",cipCode);
if(cipCode=="R"){
isR=1;
} else {
isR=2;
}
}
invariant = (invariant << 2) | isR;
}
// now deal with cis/trans - this is meant to address issue 174
// loop over the bonds on this atom and check if we have a double bond with
// a chiral code marking
if (includeChirality) {
ROMol::OBOND_ITER_PAIR atomBonds = atom->getOwningMol().getAtomBonds(atom);
int isT=0;
while (atomBonds.first != atomBonds.second){
BOND_SPTR tBond = atom->getOwningMol()[*(atomBonds.first)];
if( (tBond->getBondType() == Bond::DOUBLE) &&
(tBond->getStereo()>Bond::STEREOANY )) {
if (tBond->getStereo()==Bond::STEREOE) {
isT = 1;
} else if(tBond->getStereo()==Bond::STEREOZ) {
isT=2;
}
break;
}
atomBonds.first++;
}
invariant = (invariant << 2) | isT;
}
tres[atsSoFar++] = invariant;
}
if(includeChirality){
// ring stereochemistry
boost::dynamic_bitset<> adjusted(mol.getNumAtoms());
for(ROMol::ConstAtomIterator atIt=mol.beginAtoms();atIt!=mol.endAtoms();atIt++){
Atom const *atom = *atIt;
tres[atom->getIdx()] = tres[atom->getIdx()]<<2;
}
for(ROMol::ConstAtomIterator atIt=mol.beginAtoms();atIt!=mol.endAtoms();atIt++){
Atom const *atom = *atIt;
if((atom->getChiralTag()==Atom::CHI_TETRAHEDRAL_CW ||
atom->getChiralTag()==Atom::CHI_TETRAHEDRAL_CCW) &&
atom->hasProp("_ringStereoAtoms")){
//atom->hasProp("_CIPRank") &&
//!atom->hasProp("_CIPCode")){
ROMol::ADJ_ITER beg,end;
boost::tie(beg,end) = mol.getAtomNeighbors(atom);
unsigned int nCount=0;
while(beg!=end){
unsigned int nbrIdx=mol[*beg]->getIdx();
if(!adjusted[nbrIdx]){
tres[nbrIdx] |= nCount%4;
adjusted.set(nbrIdx);
}
++nCount;
++beg;
}
}
}
}
for(unsigned int i=0;i<mol.getNumAtoms();++i) res[i]=tres[i];
}
void buildFragmentAtomInvariants(const ROMol &mol,INVAR_VECT &res,
bool includeChirality,
const boost::dynamic_bitset<> &atomsToUse,
const boost::dynamic_bitset<> &bondsToUse,
const std::vector<std::string> *atomSymbols
){
PRECONDITION(res.size()>=mol.getNumAtoms(),"res vect too small");
std::vector<int> degrees(mol.getNumAtoms(),0);
for(unsigned int i=0;i<bondsToUse.size();++i){
if(bondsToUse[i]){
const Bond *bnd=mol.getBondWithIdx(i);
degrees[bnd->getBeginAtomIdx()]++;
degrees[bnd->getEndAtomIdx()]++;
}
}
for(ROMol::ConstAtomIterator atIt=mol.beginAtoms();atIt!=mol.endAtoms();++atIt){
Atom const *atom = *atIt;
int aIdx=atom->getIdx();
if(!atomsToUse[aIdx]){
res[aIdx] = 0;
continue;
}
boost::uint64_t invariant = 0;
int nConns = degrees[aIdx]% 8;
invariant = (invariant << 3) | nConns;
if(!atomSymbols){
int chg = abs(atom->getFormalCharge()) % 8;
int chgSign = atom->getFormalCharge() > 0;
int num = atom->getAtomicNum() % 128;
int deltaMass=0;
if(atom->getIsotope()){
deltaMass = static_cast<int>(atom->getIsotope() -
PeriodicTable::getTable()->getMostCommonIsotope(atom->getAtomicNum()));
deltaMass += 128;
if(deltaMass < 0) deltaMass = 0;
else deltaMass = deltaMass % 256;
}
invariant = (invariant << 7) | num;
invariant = (invariant << 8) | deltaMass;
invariant = (invariant << 3) | chg;
invariant = (invariant << 1) | chgSign;
invariant = (invariant << 1) | atom->getIsAromatic();
} else {
const std::string &symb=(*atomSymbols)[aIdx];
boost::uint32_t hsh=gboost::hash_range(symb.begin(),symb.end());
invariant = (invariant << 20) | (hsh%(1<<20));
}
// figure out the minimum-sized ring we're involved in
int inRing = mol.getRingInfo()->minAtomRingSize(aIdx);
inRing = inRing % 16;
invariant = (invariant << 4) | inRing;
if(includeChirality ){
int isR=0;
if( atom->hasProp("_CIPCode")){
std::string cipCode;
atom->getProp("_CIPCode",cipCode);
if(cipCode=="R"){
isR=1;
} else {
isR=2;
}
}
invariant = (invariant << 2) | isR;
}
// now deal with cis/trans - this is meant to address issue 174
// loop over the bonds on this atom and check if we have a double bond with
// a chiral code marking
if (includeChirality) {
ROMol::OBOND_ITER_PAIR atomBonds = mol.getAtomBonds(atom);
int isT=0;
while (atomBonds.first != atomBonds.second){
BOND_SPTR tBond = mol[*(atomBonds.first)];
atomBonds.first++;
if(!bondsToUse[tBond->getIdx()]) continue;
if( (tBond->getBondType() == Bond::DOUBLE) &&
(tBond->getStereo()>Bond::STEREOANY )) {
if (tBond->getStereo()==Bond::STEREOE) {
isT = 1;
} else if(tBond->getStereo()==Bond::STEREOZ) {
isT=2;
}
break;
}
}
invariant = (invariant << 2) | isT;
}
res[aIdx] = invariant;
}
}
}// end of RankAtoms namespace
namespace RDKit{
namespace MolOps {
// --------------------------------------------------
//
// Daylight canonicalization, loosely based up on algorithm described in
// JCICS 29, 97-101, (1989)
// When appropriate, specific references are made to the algorithm
// description in that paper. Steps refer to Table III of the paper
//
// --------------------------------------------------
void rankAtoms(const ROMol &mol,INT_VECT &ranks,
bool breakTies,
bool includeChirality,
bool includeIsotopes,
VECT_INT_VECT *rankHistory){
unsigned int i;
unsigned int nAtoms = mol.getNumAtoms();
PRECONDITION(ranks.size()>=nAtoms,"");
PRECONDITION(!rankHistory||rankHistory->size()>=nAtoms,"bad rankHistory size");
unsigned int stagnantTol=1;
if(!mol.getRingInfo()->isInitialized()){
MolOps::findSSSR(mol);
}
if(nAtoms > 1){
double *adjMat = MolOps::getAdjacencyMatrix(mol, true);
// ----------------------
// generate atomic invariants, Step (1)
// ----------------------
INVAR_VECT invariants;
invariants.resize(nAtoms);
RankAtoms::buildAtomInvariants(mol,invariants,includeChirality,includeIsotopes);
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "invariants:" << std::endl;
for(i=0;i<nAtoms;i++){
BOOST_LOG(rdDebugLog)<< i << " " << (long)invariants[i]<< std::endl;
}
#endif
DOUBLE_VECT atomicVect;
atomicVect.resize(nAtoms);
// ----------------------
// iteration 1: Steps (3) and (4)
// ----------------------
// Unlike the original paper, we're going to keep track of the
// ranks at each iteration and use those vectors to rank
// atoms. This seems to lead to more stable evolution of the
// ranks by avoiding ranks oscillating back and forth across
// iterations.
VECT_DOUBLE_VECT nRanks(nAtoms);
for(i=0;i<nAtoms;i++) nRanks[i].push_back(invariants[i]);
// start by ranking the atoms using the invariants
ranks.resize(nAtoms);
RankAtoms::rankVect(nRanks,ranks);
if(rankHistory){
for(i=0;i<nAtoms;i++){
(*rankHistory)[i].push_back(ranks[i]);
}
}
// how many classes are present?
unsigned int numClasses = RankAtoms::countClasses(ranks);
if(numClasses != nAtoms){
INT_VECT primeVect;
primeVect.reserve(nAtoms);
DOUBLE_VECT atomicVect;
atomicVect.reserve(nAtoms);
// indicesInPlay is used to track the atoms with non-unique ranks
// (we'll be modifying these in each step)
INT_LIST indicesInPlay;
for(i=0;i<nAtoms;i++) indicesInPlay.push_back(i);
// if we aren't breaking ties here, allow the rank iteration to
// go the full number of atoms:
if(!breakTies) stagnantTol=nAtoms;
bool done=indicesInPlay.empty();
while(!done){
//
// do one round of iterations
//
numClasses = RankAtoms::iterateRanks2(nAtoms,primeVect,atomicVect,
indicesInPlay,adjMat,ranks,nRanks,
rankHistory,stagnantTol);
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "************************ done outer iteration" << std::endl;
#endif
#ifdef VERBOSE_CANON
unsigned int tmpI;
BOOST_LOG(rdDebugLog)<< "RANKS:" << std::endl;
for(tmpI=0;tmpI<ranks.size();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\t" << tmpI << " " << ranks[tmpI] << std::endl;
}
BOOST_LOG(rdDebugLog)<< std::endl;
#endif
//
// This is the tiebreaker stage of things
//
if( breakTies && !indicesInPlay.empty() && numClasses<nAtoms){
INT_VECT newRanks = ranks;
// Add one to all ranks and multiply by two
BOOST_FOREACH(int &nr,newRanks) {
nr=(nr+1)*2;
}
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "postmult:" << std::endl;
for(tmpI=0;tmpI<newRanks.size();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\t" << newRanks[tmpI] << std::endl;
}
BOOST_LOG(rdDebugLog)<< std::endl;
#endif
//
// find lowest duplicate rank with lowest invariant:
//
int lowestIdx=indicesInPlay.front();
double lowestInvariant = invariants[lowestIdx];
int lowestRank=newRanks[lowestIdx];
BOOST_FOREACH(int ilidx,indicesInPlay){
if(newRanks[ilidx]<=lowestRank){
if(newRanks[ilidx]<lowestRank ||
invariants[ilidx] <= lowestInvariant){
lowestRank = newRanks[ilidx];
lowestIdx = ilidx;
lowestInvariant = invariants[ilidx];
}
}
}
//
// subtract one from the lowest index, rerank and proceed
//
newRanks[lowestIdx] -= 1;
RankAtoms::rankVect(newRanks,ranks);
BOOST_FOREACH(int ilidx,indicesInPlay){
nRanks[ilidx].push_back(ranks[ilidx]);
}
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "RE-RANKED ON:" << lowestIdx << std::endl;
for(tmpI=0;tmpI<newRanks.size();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\t" << newRanks[tmpI] << " " << ranks[tmpI] << std::endl;
}
BOOST_LOG(rdDebugLog)<< std::endl;
#endif
} else {
done = true;
}
}
}
}
} // end of function rankAtoms
void rankAtomsInFragment(const ROMol &mol,INT_VECT &ranks,
const boost::dynamic_bitset<> &atomsToUse,
const boost::dynamic_bitset<> &bondsToUse,
const std::vector<std::string> *atomSymbols,
const std::vector<std::string> *bondSymbols,
bool breakTies,
VECT_INT_VECT *rankHistory){
unsigned int nAtoms = mol.getNumAtoms();
unsigned int nActiveAtoms = atomsToUse.count();
PRECONDITION(ranks.size()>=nAtoms,"");
PRECONDITION(!atomSymbols||atomSymbols->size()>=nAtoms,"bad atomSymbols");
PRECONDITION(!rankHistory||rankHistory->size()>=nAtoms,"bad rankHistory size");
PRECONDITION(mol.getRingInfo()->isInitialized(),"no ring information present");
PRECONDITION(!rankHistory,"rankHistory not currently supported.");
unsigned int stagnantTol=1;
if(nActiveAtoms > 1){
// ----------------------
// generate atomic invariants, Step (1)
// ----------------------
INVAR_VECT invariants;
invariants.resize(nAtoms);
RankAtoms::buildFragmentAtomInvariants(mol,invariants,true,
atomsToUse,bondsToUse,
atomSymbols);
INVAR_VECT tinvariants;
tinvariants.resize(nActiveAtoms);
unsigned int activeIdx=0;
for(unsigned int aidx=0;aidx<nAtoms;++aidx){
if(atomsToUse[aidx]){
tinvariants[activeIdx++]=invariants[aidx];
}
}
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "invariants:" << std::endl;
for(unsigned int i=0;i<nActiveAtoms;i++){
BOOST_LOG(rdDebugLog)<< i << " " << (long)tinvariants[i]<< std::endl;
}
#endif
// ----------------------
// iteration 1: Steps (3) and (4)
// ----------------------
// start by ranking the atoms using the invariants
VECT_DOUBLE_VECT nRanks(nActiveAtoms);
for(unsigned int i=0;i<nActiveAtoms;i++) nRanks[i].push_back(tinvariants[i]);
INT_VECT tranks(nActiveAtoms,0);
RankAtoms::rankVect(nRanks,tranks);
#if 0
if(rankHistory){
for(unsigned int i=0;i<nAtoms;i++){
(*rankHistory)[i].push_back(ranks[i]);
}
}
#endif
// how many classes are present?
unsigned int numClasses = RankAtoms::countClasses(tranks);
if(numClasses != nActiveAtoms){
double *tadjMat = new double[nActiveAtoms*nActiveAtoms];
memset(static_cast<void *>(tadjMat),0,nActiveAtoms*nActiveAtoms*sizeof(double));
if(!bondSymbols){
double *adjMat = MolOps::getAdjacencyMatrix(mol,true,0,true,0,&bondsToUse);
activeIdx=0;
for(unsigned int aidx=0;aidx<nAtoms;++aidx){
if(atomsToUse[aidx]){
unsigned int activeIdx2=activeIdx+1;
for(unsigned int aidx2=aidx+1;aidx2<nAtoms;++aidx2){
if(atomsToUse[aidx2]){
tadjMat[activeIdx*nActiveAtoms+activeIdx2]=adjMat[aidx*nAtoms+aidx2];
tadjMat[activeIdx2*nActiveAtoms+activeIdx]=adjMat[aidx2*nAtoms+aidx];
++activeIdx2;
}
}
++activeIdx;
}
}
} else {
// rank the bond symbols we have:
std::vector<boost::uint32_t> tbranks(bondsToUse.size(),
0);
for(unsigned int bidx=0;bidx<bondsToUse.size();++bidx){
if(!bondsToUse[bidx]) continue;
const std::string &symb=(*bondSymbols)[bidx];
boost::uint32_t hsh=gboost::hash_range(symb.begin(),symb.end());
tbranks[bidx]=hsh;
}
INT_VECT branks(bondsToUse.size(),1000000);
#ifdef VERBOSE_CANON
std::cerr<<" tbranks:";
std::copy(tbranks.begin(),tbranks.end(),std::ostream_iterator<boost::uint32_t>(std::cerr," "));
std::cerr<<std::endl;
#endif
RankAtoms::rankVect(tbranks,branks);
#ifdef VERBOSE_CANON
std::cerr<<" branks:";
std::copy(branks.begin(),branks.end(),std::ostream_iterator<int>(std::cerr," "));
std::cerr<<std::endl;
#endif
for(unsigned int bidx=0;bidx<bondsToUse.size();++bidx){
if(!bondsToUse[bidx]) continue;
const Bond *bond=mol.getBondWithIdx(bidx);
unsigned int aidx1=bond->getBeginAtomIdx();
unsigned int aidx2=bond->getEndAtomIdx();
unsigned int tidx1=0;
for(unsigned int iidx=0;iidx<aidx1;++iidx){
if(atomsToUse[iidx]) ++tidx1;
}
unsigned int tidx2=0;
for(unsigned int iidx=0;iidx<aidx2;++iidx){
if(atomsToUse[iidx]) ++tidx2;
}
//const std::string &symb=(*bondSymbols)[bidx];
//boost::uint32_t hsh=gboost::hash_range(symb.begin(),symb.end());
//std::cerr<<" ::: "<<bidx<<"->"<<branks[bidx]<<std::endl;
tadjMat[tidx1*nActiveAtoms+tidx2]=branks[bidx];
tadjMat[tidx2*nActiveAtoms+tidx1]=branks[bidx];
}
}
INT_VECT primeVect;
primeVect.reserve(nActiveAtoms);
DOUBLE_VECT atomicVect;
atomicVect.reserve(nActiveAtoms);
#ifdef VERBOSE_CANON
for(unsigned int aidx1=0;aidx1<nActiveAtoms;++aidx1){
std::cerr<<aidx1<<" : ";
for(unsigned int aidx2=aidx1+1;aidx2<nActiveAtoms;++aidx2){
std::cerr<< tadjMat[aidx1*nActiveAtoms+aidx2]<<" ";
}
std::cerr<<std::endl;
}
#endif
// indicesInPlay is used to track the atoms with non-unique ranks
// (we'll be modifying these in each step)
INT_LIST indicesInPlay;
for(unsigned int i=0;i<nActiveAtoms;i++) indicesInPlay.push_back(i);
// if we aren't breaking ties here, allow the rank iteration to
// go the full number of atoms:
if(!breakTies) stagnantTol=nActiveAtoms;
bool done=indicesInPlay.empty();
while(!done){
//
// do one round of iterations
//
numClasses = RankAtoms::iterateRanks2(nActiveAtoms,primeVect,atomicVect,
indicesInPlay,tadjMat,tranks,nRanks,
rankHistory,stagnantTol);
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "************************ done outer iteration" << std::endl;
#endif
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "RANKS:" << std::endl;
for(unsigned int tmpI=0;tmpI<tranks.size();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\t" << tmpI << " " << tranks[tmpI] << std::endl;
}
BOOST_LOG(rdDebugLog)<< std::endl;
#endif
//
// This is the tiebreaker stage of things
//
if( breakTies && !indicesInPlay.empty() && numClasses<nActiveAtoms){
INT_VECT newRanks = tranks;
// Add one to all ranks and multiply by two
std::for_each(newRanks.begin(),newRanks.end(),_1=(_1+1)*2);
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "postmult:" << std::endl;
for(unsigned tmpI=0;tmpI<newRanks.size();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\t" << newRanks[tmpI] << std::endl;
}
BOOST_LOG(rdDebugLog)<< std::endl;
#endif
//
// find lowest duplicate rank with lowest invariant:
//
int lowestIdx=indicesInPlay.front();
double lowestInvariant = tinvariants[lowestIdx];
int lowestRank=newRanks[lowestIdx];
for(INT_LIST_I ilIt=indicesInPlay.begin();
ilIt!=indicesInPlay.end();
++ilIt){
if(newRanks[*ilIt]<=lowestRank){
if(newRanks[*ilIt]<lowestRank ||
tinvariants[*ilIt] <= lowestInvariant){
lowestRank = newRanks[*ilIt];
lowestIdx = *ilIt;
lowestInvariant = tinvariants[*ilIt];
}
}
}
//
// subtract one from the lowest index, rerank and proceed
//
newRanks[lowestIdx] -= 1;
RankAtoms::rankVect(newRanks,tranks);
BOOST_FOREACH(int ilidx,indicesInPlay){
nRanks[ilidx].push_back(ranks[ilidx]);
}
#ifdef VERBOSE_CANON
BOOST_LOG(rdDebugLog)<< "RE-RANKED ON:" << lowestIdx << std::endl;
for(unsigned int tmpI=0;tmpI<newRanks.size();tmpI++){
BOOST_LOG(rdDebugLog)<< "\t\t" << newRanks[tmpI] << " " << tranks[tmpI] << std::endl;
}
BOOST_LOG(rdDebugLog)<< std::endl;
#endif
} else {
done = true;
}
}
delete [] tadjMat;
}
unsigned int tidx=0;
for(unsigned int aidx=0;aidx<nAtoms;++aidx){
ranks[aidx]=0;
if(atomsToUse[aidx]){
ranks[aidx]=tranks[tidx++];
}
}
}
} // end of function rankAtomsInFragment
} // end of namespace MolOps
} // End Of RDKit namespace
|
