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 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449
|
//
// Copyright (C) 2014 Novartis Institutes for BioMedical Research
//
// @@ 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 <list>
#include <algorithm>
#include <cmath>
#include "../QueryAtom.h"
#include "../QueryBond.h"
#include "../SmilesParse/SmilesWrite.h"
#include "../SmilesParse/SmartsWrite.h"
#include "../SmilesParse/SmilesParse.h"
#include "../Substruct/SubstructMatch.h"
#include "SubstructMatchCustom.h"
#include "MaximumCommonSubgraph.h"
#include <RDGeneral/BoostStartInclude.h>
#include <boost/graph/adjacency_list.hpp>
#include <boost/dynamic_bitset.hpp>
#include <RDGeneral/BoostEndInclude.h>
namespace RDKit {
namespace FMCS {
struct LabelDefinition {
unsigned int ItemIndex; // item with this label value
unsigned int Value;
LabelDefinition() : ItemIndex(NotSet), Value(NotSet) {}
LabelDefinition(unsigned int i, unsigned int value)
: ItemIndex(i), Value(value) {}
};
MaximumCommonSubgraph::MaximumCommonSubgraph(const MCSParameters *params) {
Parameters = detail::MCSParametersInternal(params);
if (!Parameters.ProgressCallback) {
Parameters.ProgressCallback = MCSProgressCallbackTimeout;
Parameters.ProgressCallbackUserData = &To;
}
if (Parameters.AtomCompareParameters.MatchChiralTag) {
Parameters.BondCompareParameters.MatchStereo = true;
}
QueryMoleculeMatchedBonds = 0;
QueryMoleculeMatchedAtoms = 0;
QueryMoleculeSingleMatchedAtom = nullptr;
To = nanoClock();
}
static bool molPtr_NumBondLess(
const ROMol *l,
const ROMol *r) { // need for sorting the source molecules by size
return l->getNumBonds() < r->getNumBonds();
}
void MaximumCommonSubgraph::init(size_t startIdx) {
QueryMolecule = Molecules.at(startIdx);
Targets.clear();
#ifdef FAST_SUBSTRUCT_CACHE
QueryAtomLabels.clear();
QueryBondLabels.clear();
QueryAtomMatchTable.clear();
QueryBondMatchTable.clear();
#endif
#ifdef DUP_SUBSTRUCT_CACHE
DuplicateCache.clear();
#endif
size_t nq = 0;
#ifdef FAST_SUBSTRUCT_CACHE
// fill out match tables
nq = QueryMolecule->getNumAtoms();
QueryAtomMatchTable.resize(nq, nq);
for (size_t aj = 0; aj < nq; aj++) {
for (size_t ai = 0; ai < nq; ai++) {
QueryAtomMatchTable.set(
ai, aj,
Parameters.AtomTyper(Parameters.AtomCompareParameters, *QueryMolecule,
ai, *QueryMolecule, aj,
Parameters.CompareFunctionsUserData));
}
}
nq = QueryMolecule->getNumBonds();
QueryBondMatchTable.resize(nq, nq);
for (size_t aj = 0; aj < nq; aj++) {
for (size_t ai = 0; ai < nq; ai++) {
QueryBondMatchTable.set(
ai, aj,
Parameters.BondTyper(Parameters.BondCompareParameters, *QueryMolecule,
ai, *QueryMolecule, aj,
Parameters.CompareFunctionsUserData));
}
}
// Compute label values based on current functor and parameters for code
// Morgan correct computation.
unsigned int currentLabelValue = 1;
std::vector<LabelDefinition> labels;
nq = QueryMolecule->getNumAtoms();
QueryAtomLabels.resize(nq, NotSet);
for (size_t ai = 0; ai < nq; ++ai) {
if (MCSAtomCompareAny ==
Parameters.AtomTyper) { // predefined functor without atom compare
// parameters
QueryAtomLabels[ai] = 1;
} else {
const auto atom = QueryMolecule->getAtomWithIdx(ai);
if (MCSAtomCompareElements ==
Parameters.AtomTyper) { // predefined functor without atom compare
// parameters
QueryAtomLabels[ai] = atom->getAtomicNum() |
(Parameters.AtomCompareParameters.MatchValences
? (atom->getTotalValence() >> 8)
: 0);
} else if (MCSAtomCompareIsotopes ==
Parameters.AtomTyper) { // predefined functor without atom
// compare parameters
QueryAtomLabels[ai] = atom->getAtomicNum() | (atom->getIsotope() >> 8) |
(Parameters.AtomCompareParameters.MatchValences
? (atom->getTotalValence() >> 16)
: 0);
} else { // custom user defined functor
QueryAtomLabels[ai] = NotSet;
for (auto &label : labels) {
if (Parameters.AtomTyper(
Parameters.AtomCompareParameters, *QueryMolecule,
label.ItemIndex, *QueryMolecule, ai,
Parameters.CompareFunctionsUserData)) { // equal items
QueryAtomLabels[ai] = label.Value;
break;
}
}
if (NotSet ==
QueryAtomLabels.at(ai)) { // not found -> create new label
QueryAtomLabels[ai] = ++currentLabelValue;
labels.emplace_back(ai, currentLabelValue);
}
}
}
}
labels.clear();
currentLabelValue = 1;
nq = QueryMolecule->getNumBonds();
QueryBondLabels.resize(nq, NotSet);
for (size_t aj = 0; aj < nq; ++aj) {
const Bond *bond = QueryMolecule->getBondWithIdx(aj);
unsigned ring = 0;
if (!Parameters.CompareFunctionsUserData &&
(Parameters.BondCompareParameters.CompleteRingsOnly ||
Parameters.BondCompareParameters.RingMatchesRingOnly)) {
// is bond in ring
ring = QueryMolecule->getRingInfo()->numBondRings(aj) ? 0 : 1;
}
if (MCSBondCompareAny == Parameters.BondTyper) {
QueryBondLabels[aj] = 1 | (ring >> 8);
} else if (MCSBondCompareOrderExact == Parameters.BondTyper) {
QueryBondLabels[aj] = (bond->getBondType() + 1) | (ring >> 8);
} else if (MCSBondCompareOrder == Parameters.BondTyper) {
auto order = bond->getBondType();
if (Bond::AROMATIC == order ||
Bond::ONEANDAHALF == order) { // ignore Aromatization
order = Bond::SINGLE;
} else if (Bond::TWOANDAHALF == order) {
order = Bond::DOUBLE;
} else if (Bond::THREEANDAHALF == order) {
order = Bond::TRIPLE;
} else if (Bond::FOURANDAHALF == order) {
order = Bond::QUADRUPLE;
} else if (Bond::FIVEANDAHALF == order) {
order = Bond::QUINTUPLE;
}
QueryBondLabels[aj] = (order + 1) | (ring >> 8);
} else { // custom user defined functor
QueryBondLabels[aj] = NotSet;
for (const auto &label : labels) {
if (Parameters.BondTyper(
Parameters.BondCompareParameters, *QueryMolecule,
label.ItemIndex, *QueryMolecule, aj,
Parameters
.CompareFunctionsUserData)) { // equal bonds + ring ...
QueryBondLabels[aj] = label.Value;
break;
}
}
if (NotSet == QueryBondLabels.at(aj)) { // not found -> create new label
QueryBondLabels[aj] = ++currentLabelValue;
labels.emplace_back(aj, currentLabelValue);
}
}
}
#endif
Targets.resize(Molecules.size() - 1);
size_t i = 0;
for (auto it = Molecules.begin() + 1; it != Molecules.end(); it++, i++) {
Targets[i].Molecule = *it;
// build Target Topology ADD ATOMs
for (const auto &a : Targets.at(i).Molecule->atoms()) {
Targets[i].Topology.addAtom(a->getIdx());
}
// build Target Topology ADD BONDs
for (const auto &b : Targets.at(i).Molecule->bonds()) {
auto ii = b->getBeginAtomIdx();
auto jj = b->getEndAtomIdx();
Targets[i].Topology.addBond(b->getIdx(), ii, jj);
}
// fill out match tables
size_t nq = QueryMolecule->getNumAtoms();
size_t nt = Targets.at(i).Molecule->getNumAtoms();
Targets[i].AtomMatchTable.resize(nq, nt);
for (size_t aj = 0; aj < nt; aj++) {
for (size_t ai = 0; ai < nq; ai++) {
Targets[i].AtomMatchTable.set(
ai, aj,
Parameters.AtomTyper(Parameters.AtomCompareParameters,
*QueryMolecule, ai, *Targets.at(i).Molecule,
aj, Parameters.CompareFunctionsUserData));
}
}
nq = QueryMolecule->getNumBonds();
nt = Targets.at(i).Molecule->getNumBonds();
Targets[i].BondMatchTable.resize(nq, nt);
for (size_t aj = 0; aj < nt; aj++) {
for (size_t ai = 0; ai < nq; ai++) {
Targets[i].BondMatchTable.set(
ai, aj,
Parameters.BondTyper(Parameters.BondCompareParameters,
*QueryMolecule, ai, *Targets.at(i).Molecule,
aj, Parameters.CompareFunctionsUserData));
}
}
}
}
struct WeightedBond {
const Bond *BondPtr{nullptr};
unsigned int Weight{0};
WeightedBond() {}
WeightedBond(const Bond *bond) : BondPtr(bond), Weight(0) {
const auto ringInfo = bond->getOwningMol().getRingInfo();
// score ((bond.is_in_ring + atom1.is_in_ring + atom2.is_in_ring)
if (ringInfo->numBondRings(bond->getIdx())) {
++Weight;
}
if (ringInfo->numAtomRings(bond->getBeginAtomIdx())) {
++Weight;
}
if (ringInfo->numAtomRings(bond->getEndAtomIdx())) {
++Weight;
}
}
bool operator<(const WeightedBond &r) {
return Weight >= r.Weight; // sort in Z-A order (Rings first)
}
};
void MaximumCommonSubgraph::makeInitialSeeds() {
// build a set of initial seeds as "all" single bonds from query
// molecule
boost::dynamic_bitset<> excludedBonds(QueryMolecule->getNumBonds());
Seeds.clear();
QueryMoleculeMatchedBonds = 0;
QueryMoleculeMatchedAtoms = 0;
QueryMoleculeSingleMatchedAtom = nullptr;
if (!Parameters.InitialSeed.empty()) { // make user defined seed
std::unique_ptr<const ROMol> initialSeedMolecule(
static_cast<const ROMol *>(SmartsToMol(Parameters.InitialSeed)));
// make a set of of seed as indices and pointers to current query
// molecule items based on matching results
std::vector<MatchVectType> matching_substructs;
SubstructMatch(*QueryMolecule, *initialSeedMolecule, matching_substructs);
// loop throw all fragments of Query matched to initial seed
for (const auto &ms : matching_substructs) {
Seed seed;
seed.setStoreAllDegenerateMCS(Parameters.StoreAll);
seed.ExcludedBonds = excludedBonds;
seed.MatchResult.resize(Targets.size());
#ifdef VERBOSE_STATISTICS_ON
{
++VerboseStatistics.Seed;
++VerboseStatistics.InitialSeed;
}
#endif
// add all matched atoms of the matched query fragment
std::map<unsigned int, unsigned int> initialSeedToQueryAtom;
for (const auto &msb : ms) {
unsigned int qai = msb.second;
unsigned int sai = msb.first;
seed.addAtom(QueryMolecule->getAtomWithIdx(qai));
initialSeedToQueryAtom[sai] = qai;
}
// add all bonds (existed in initial seed !!!) between all matched
// atoms in query
for (const auto &msb : ms) {
const auto atom = initialSeedMolecule->getAtomWithIdx(msb.first);
for (const auto &nbri : boost::make_iterator_range(
initialSeedMolecule->getAtomBonds(atom))) {
const auto initialBond = (*initialSeedMolecule)[nbri];
unsigned int qai1 =
initialSeedToQueryAtom.at(initialBond->getBeginAtomIdx());
unsigned int qai2 =
initialSeedToQueryAtom.at(initialBond->getEndAtomIdx());
const auto b = QueryMolecule->getBondBetweenAtoms(qai1, qai2);
CHECK_INVARIANT(b, "bond must not be NULL");
if (!seed.ExcludedBonds.test(b->getIdx())) {
seed.addBond(b);
seed.ExcludedBonds.set(b->getIdx());
}
}
}
seed.computeRemainingSize(*QueryMolecule);
if (checkIfMatchAndAppend(seed)) {
QueryMoleculeMatchedBonds = seed.getNumBonds();
}
}
if (Seeds.empty()) {
BOOST_LOG(rdWarningLog)
<< "The provided InitialSeed is not an MCS and will be ignored"
<< std::endl;
}
}
if (Seeds.empty()) { // create a set of seeds from each query bond
// R1 additional performance OPTIMISATION
// if(Parameters.BondCompareParameters.CompleteRingsOnly)
// disable all mismatched rings, and do not generate initial seeds
// from such disabled bonds
// for( rings .....) for(i......)
// if(mismatched) excludedBonds[i.......] = true;
std::vector<WeightedBond> wbVec;
wbVec.reserve(QueryMolecule->getNumBonds());
for (const auto &bond : QueryMolecule->bonds()) {
wbVec.emplace_back(bond);
}
for (const auto &wb : wbVec) {
// R1 additional performance OPTIMISATION
// if(excludedBonds[(*bi)->getIdx()])
// continue;
Seed seed;
seed.setStoreAllDegenerateMCS(Parameters.StoreAll);
seed.MatchResult.resize(Targets.size());
#ifdef VERBOSE_STATISTICS_ON
{
++VerboseStatistics.Seed;
++VerboseStatistics.InitialSeed;
}
#endif
seed.addAtom(wb.BondPtr->getBeginAtom());
seed.addAtom(wb.BondPtr->getEndAtom());
seed.ExcludedBonds = excludedBonds; // all bonds from first to current
seed.addBond(wb.BondPtr);
excludedBonds.set(wb.BondPtr->getIdx());
seed.computeRemainingSize(*QueryMolecule);
if (checkIfMatchAndAppend(seed)) {
++QueryMoleculeMatchedBonds;
} else {
// optionally remove all such bonds from all targets TOPOLOGY
// where it exists.
//..........
// disable (mark as already processed) mismatched bond in all
// seeds
for (auto &Seed : Seeds) {
Seed.ExcludedBonds.set(wb.BondPtr->getIdx());
}
#ifdef VERBOSE_STATISTICS_ON
++VerboseStatistics.MismatchedInitialSeed;
#endif
}
}
}
auto nq = QueryMolecule->getNumAtoms();
MatchVectType singleAtomPairMatch(1);
MatchVectType emptyBondPairMatch;
for (size_t i = 0; i < nq; i++) { // all query's atoms
const auto queryMolAtom = QueryMolecule->getAtomWithIdx(i);
bool isQueryMolAtomInRing = queryIsAtomInRing(queryMolAtom);
unsigned int matched = 0;
const Atom *candQueryMoleculeSingleMatchedAtom = nullptr;
for (const auto &tag : Targets) {
auto nt = tag.Molecule->getNumAtoms();
for (size_t aj = 0; aj < nt; aj++) {
if (tag.AtomMatchTable.at(i, aj)) {
const auto targetMolAtom = tag.Molecule->getAtomWithIdx(aj);
bool isTargetMolAtomInRing = queryIsAtomInRing(targetMolAtom);
++matched;
if (!(Parameters.BondCompareParameters.CompleteRingsOnly &&
(isQueryMolAtomInRing || isTargetMolAtomInRing))) {
bool shouldAccept = !Parameters.ShouldAcceptMCS;
if (!shouldAccept) {
singleAtomPairMatch[0] = std::make_pair(i, aj);
shouldAccept = Parameters.ShouldAcceptMCS(
*QueryMolecule, *tag.Molecule, singleAtomPairMatch,
emptyBondPairMatch, &Parameters);
}
if (shouldAccept) {
candQueryMoleculeSingleMatchedAtom = queryMolAtom;
}
}
break;
}
}
}
if (matched && matched >= ThresholdCount) {
++QueryMoleculeMatchedAtoms;
if (candQueryMoleculeSingleMatchedAtom) {
if (!QueryMoleculeSingleMatchedAtom) {
QueryMoleculeSingleMatchedAtom = candQueryMoleculeSingleMatchedAtom;
} else {
QueryMoleculeSingleMatchedAtom = (std::max)(
candQueryMoleculeSingleMatchedAtom,
QueryMoleculeSingleMatchedAtom, [](const Atom *a, const Atom *b) {
if (a->getDegree() != b->getDegree()) {
return (a->getDegree() < b->getDegree());
} else if (a->getFormalCharge() != b->getFormalCharge()) {
return (a->getFormalCharge() < b->getFormalCharge());
} else if (a->getAtomicNum() != b->getAtomicNum()) {
return (a->getAtomicNum() < b->getAtomicNum());
}
return (a->getIdx() < b->getIdx());
});
}
}
}
}
}
namespace {
bool checkIfRingsAreClosed(const Seed &fs, bool noLoneRingAtoms) {
if (fs.MoleculeFragment.Bonds.empty() && fs.MoleculeFragment.Atoms.empty()) {
return true;
}
const auto &om = fs.MoleculeFragment.Atoms.front()->getOwningMol();
const auto ri = om.getRingInfo();
if (!ri->numRings()) {
return true;
}
boost::dynamic_bitset<> mcsBonds(om.getNumBonds());
boost::dynamic_bitset<> mcsNonFusedRings(ri->numRings());
boost::dynamic_bitset<> mcsFusedRings(ri->numRings());
for (const auto &bond : fs.MoleculeFragment.Bonds) {
auto bi = bond->getIdx();
mcsBonds.set(bi);
if (ri->numBondRings(bi) == 1) {
mcsNonFusedRings.set(ri->bondMembers(bi).front());
}
}
for (unsigned int ringIdx = 0; ringIdx < mcsNonFusedRings.size(); ++ringIdx) {
if (!mcsNonFusedRings.test(ringIdx)) {
continue;
}
for (const auto &bi : ri->bondRings().at(ringIdx)) {
bool keepBond = false;
for (unsigned int memberOf : ri->bondMembers(bi)) {
if (memberOf == ringIdx) {
keepBond = true;
} else if (mcsNonFusedRings.test(memberOf)) {
keepBond = false;
break;
}
}
if (keepBond && !mcsBonds.test(bi)) {
return false;
}
}
}
if (noLoneRingAtoms) {
for (const auto &atom : fs.MoleculeFragment.Atoms) {
auto ai = atom->getIdx();
const auto &ringIndices = ri->atomMembers(ai);
if (!ringIndices.empty() &&
!std::any_of(ringIndices.begin(), ringIndices.end(),
[&mcsNonFusedRings](const auto &ringIdx) {
return mcsNonFusedRings.test(ringIdx);
})) {
return false;
}
}
}
if (mcsNonFusedRings.none()) {
for (const auto &bond : fs.MoleculeFragment.Bonds) {
auto bi = bond->getIdx();
if (ri->numBondRings(bi) > 1) {
for (auto ringIdx : ri->bondMembers(bi)) {
mcsFusedRings.set(ringIdx);
}
}
}
}
if (mcsFusedRings.any()) {
for (unsigned int ringIdx = 0; ringIdx < mcsFusedRings.size(); ++ringIdx) {
if (!mcsFusedRings.test(ringIdx)) {
continue;
}
const auto &ringBondIndices = ri->bondRings().at(ringIdx);
if (std::all_of(
ringBondIndices.begin(), ringBondIndices.end(),
[&mcsBonds](const auto &bi) { return mcsBonds.test(bi); })) {
return true;
}
}
return false;
}
return true;
}
bool checkIfShouldAcceptMCS(const FMCS::MolFragment &f, const ROMol &query,
const std::vector<Target> &targets,
const MCSParameters &p) {
if (!p.ShouldAcceptMCS || f.Bonds.empty()) {
return true;
}
Seed seed; // result MCS
seed.ExcludedBonds.resize(query.getNumBonds(), false);
for (const auto &atom : f.Atoms) {
seed.addAtom(atom);
}
for (const auto &bond : f.Bonds) {
seed.addBond(bond);
}
for (const auto &target : targets) {
match_V_t match;
bool targetMatched = SubstructMatchCustomTable(
target.Topology, *target.Molecule, seed.Topology, query,
target.AtomMatchTable, target.BondMatchTable, &p, &match);
// it is OK to skip non-matches here as threshold could be < 1.0;
// we only want to triage matches
if (!targetMatched) {
continue;
}
MatchVectType atomPairMatch;
atomPairMatch.reserve(match.size());
std::vector<unsigned int> queryToTargetAtomIdx(query.getNumAtoms(), NotSet);
for (const auto &m : match) {
auto queryAtomIdx = seed.Topology[m.first];
auto targetAtomIdx = target.Topology[m.second];
atomPairMatch.emplace_back(queryAtomIdx, targetAtomIdx);
queryToTargetAtomIdx[queryAtomIdx] = targetAtomIdx;
}
MatchVectType bondPairMatch;
auto numMcsBonds = boost::num_edges(seed.Topology);
bondPairMatch.reserve(numMcsBonds);
for (const auto &it :
boost::make_iterator_range(boost::edges(seed.Topology))) {
int queryBeginAtomIdx = seed.Topology[boost::source(it, seed.Topology)];
int queryEndAtomIdx = seed.Topology[boost::target(it, seed.Topology)];
const auto queryBond =
query.getBondBetweenAtoms(queryBeginAtomIdx, queryEndAtomIdx);
CHECK_INVARIANT(queryBond, "");
const auto targetBeginAtomIdx =
queryToTargetAtomIdx.at(queryBeginAtomIdx);
CHECK_INVARIANT(targetBeginAtomIdx != NotSet, "");
const auto targetEndAtomIdx = queryToTargetAtomIdx.at(queryEndAtomIdx);
CHECK_INVARIANT(targetEndAtomIdx != NotSet, "");
const auto targetBond = target.Molecule->getBondBetweenAtoms(
targetBeginAtomIdx, targetEndAtomIdx);
CHECK_INVARIANT(targetBond, "");
bondPairMatch.emplace_back(queryBond->getIdx(), targetBond->getIdx());
}
if (!p.ShouldAcceptMCS(query, *target.Molecule, atomPairMatch,
bondPairMatch, &p)) {
return false;
}
}
return true;
}
} // namespace
bool MaximumCommonSubgraph::growSeeds() {
bool mcsFound = false;
bool canceled = false;
// Find MCS -- SDF Seed growing OPTIMISATION (it works in 3 times
// faster)
while (!Seeds.empty()) {
if (getMaxNumberBonds() == QueryMoleculeMatchedBonds) { // MCS == Query
break;
}
#ifdef VERBOSE_STATISTICS_ON
VerboseStatistics.TotalSteps++;
#endif
auto si = Seeds.begin();
si->grow(*this);
{
const Seed &fs = Seeds.front();
// bigger substructure found
if (fs.CopyComplete) {
bool possibleMCS = false;
if (!Parameters.MaximizeBonds) {
possibleMCS = (fs.getNumAtoms() > getMaxNumberAtoms() ||
(fs.getNumAtoms() == getMaxNumberAtoms() &&
fs.getNumBonds() > getMaxNumberBonds()));
} else {
possibleMCS = (fs.getNumBonds() > getMaxNumberBonds() ||
(fs.getNumBonds() == getMaxNumberBonds() &&
fs.getNumAtoms() > getMaxNumberAtoms()));
}
bool isDegenerateMCS = (fs.getNumBonds() == getMaxNumberBonds() &&
fs.getNumAtoms() == getMaxNumberAtoms());
if (!possibleMCS && Parameters.StoreAll) {
possibleMCS = isDegenerateMCS;
}
// #945: test here to see if the MCS actually has all rings closed
if (possibleMCS && Parameters.BondCompareParameters.CompleteRingsOnly) {
possibleMCS = checkIfRingsAreClosed(
fs, Parameters.AtomCompareParameters.CompleteRingsOnly);
}
if (possibleMCS) {
possibleMCS = checkIfShouldAcceptMCS(
fs.MoleculeFragment, *QueryMolecule, Targets, Parameters);
}
if (possibleMCS) {
mcsFound = true;
#ifdef VERBOSE_STATISTICS_ON
VerboseStatistics.MCSFoundStep = VerboseStatistics.TotalSteps;
VerboseStatistics.MCSFoundTime = nanoClock();
#endif
McsIdx.Atoms = fs.MoleculeFragment.Atoms;
McsIdx.Bonds = fs.MoleculeFragment.Bonds;
if (Parameters.Verbose) {
std::cout << VerboseStatistics.TotalSteps
<< " Seeds:" << Seeds.size() << " MCS "
<< McsIdx.Atoms.size() << " atoms, "
<< McsIdx.Bonds.size() << " bonds";
printf(" for %.4lf seconds. bond[0]=%u\n",
double(VerboseStatistics.MCSFoundTime - To) / 1000000.,
McsIdx.Bonds.front()->getIdx());
}
if (Parameters.StoreAll) {
if (!isDegenerateMCS) {
DegenerateMcsMap.clear();
}
std::vector<unsigned int> key(McsIdx.Bonds.size());
std::transform(McsIdx.Bonds.begin(), McsIdx.Bonds.end(),
key.begin(),
[](const auto bond) { return bond->getIdx(); });
std::sort(key.begin(), key.end());
MCS value(McsIdx);
value.QueryMolecule = QueryMolecule;
value.Targets = Targets;
DegenerateMcsMap[key] = value;
}
}
}
}
if (NotSet == si->GrowingStage) { // finished
Seeds.erase(si);
}
if (Parameters.ProgressCallback) {
Stat.NumAtoms = getMaxNumberAtoms();
Stat.NumBonds = getMaxNumberBonds();
if (!Parameters.ProgressCallback(Stat, Parameters,
Parameters.ProgressCallbackUserData)) {
canceled = true;
break;
}
}
}
if (mcsFound) { // postponed copy of current set of molecules for
// threshold < 1.
McsIdx.QueryMolecule = QueryMolecule;
McsIdx.Targets = Targets;
}
return !canceled;
} // namespace FMCS
struct AtomMatch { // for each seed atom (matched)
unsigned int QueryAtomIdx;
unsigned int TargetAtomIdx;
AtomMatch() : QueryAtomIdx(NotSet), TargetAtomIdx(NotSet) {}
};
typedef std::vector<AtomMatch> AtomMatchSet;
std::pair<std::string, ROMOL_SPTR>
MaximumCommonSubgraph::generateResultSMARTSAndQueryMol(
const MCS &mcsIdx) const {
// match the result MCS with all targets to check if it is exact match
// or template
Seed seed; // result MCS
seed.setStoreAllDegenerateMCS(Parameters.StoreAll);
seed.ExcludedBonds.resize(mcsIdx.QueryMolecule->getNumBonds(), false);
std::vector<AtomMatchSet> atomMatchResult(mcsIdx.Targets.size());
std::vector<unsigned int> atomIdxMap(mcsIdx.QueryMolecule->getNumAtoms());
std::vector<std::map<unsigned int, const Bond *>> bondMatchSet(
mcsIdx.Bonds.size()); // key is unique BondType
std::vector<std::map<unsigned int, const Atom *>> atomMatchSet(
mcsIdx.Atoms.size()); // key is unique atomic number
for (const auto &atom : mcsIdx.Atoms) {
atomIdxMap[atom->getIdx()] = seed.getNumAtoms();
seed.addAtom(atom);
}
for (const auto &bond : mcsIdx.Bonds) {
seed.addBond(bond);
}
std::vector<unsigned int> matchedTargetIndices;
if (!mcsIdx.Bonds.empty()) {
for (const auto &tag : mcsIdx.Targets) {
match_V_t match; // THERE IS NO Bonds match INFO !!!!
bool target_matched = SubstructMatchCustomTable(
tag.Topology, *tag.Molecule, seed.Topology, *QueryMolecule,
tag.AtomMatchTable, tag.BondMatchTable, &Parameters, &match);
if (!target_matched) {
continue;
}
unsigned int itarget = &tag - &mcsIdx.Targets.front();
matchedTargetIndices.push_back(itarget);
atomMatchResult.at(itarget).resize(seed.getNumAtoms());
for (const auto &m : match) {
const auto ai = m.first; // SeedAtomIdx
atomMatchResult.at(itarget).at(ai).QueryAtomIdx =
seed.Topology[m.first];
atomMatchResult.at(itarget).at(ai).TargetAtomIdx =
tag.Topology[m.second];
const auto ta = tag.Molecule->getAtomWithIdx(tag.Topology[m.second]);
if (ta->getAtomicNum() !=
seed.MoleculeFragment.Atoms.at(ai)->getAtomicNum()) {
atomMatchSet[ai][ta->getAtomicNum()] = ta; // add
}
}
// AND BUILD BOND MATCH INFO
for (const auto &bond : mcsIdx.Bonds) {
const auto bi = &bond - &mcsIdx.Bonds.front();
const auto i = atomIdxMap.at(bond->getBeginAtomIdx());
const auto j = atomIdxMap.at(bond->getEndAtomIdx());
const auto ti = atomMatchResult.at(itarget).at(i).TargetAtomIdx;
const auto tj = atomMatchResult.at(itarget).at(j).TargetAtomIdx;
const auto tb = tag.Molecule->getBondBetweenAtoms(ti, tj);
if (tb && bond->getBondType() != tb->getBondType()) {
bondMatchSet[bi][tb->getBondType()] = tb; // add
}
}
}
}
// Generate result's SMARTS
// create molecule from MCS for MolToSmarts()
auto mol = new RWMol();
ROMOL_SPTR molSptr(mol);
const auto ri = mcsIdx.QueryMolecule->getRingInfo();
boost::dynamic_bitset<> mcsRingIsComplete;
bool needAtomRingQueries =
(Parameters.AtomCompareParameters.RingMatchesRingOnly ||
Parameters.BondCompareParameters.MatchFusedRingsStrict);
if (needAtomRingQueries) {
mcsRingIsComplete.resize(ri->numRings(), true);
boost::dynamic_bitset<> queryBondInMcs(mcsIdx.QueryMolecule->getNumBonds());
for (const auto &bond : mcsIdx.Bonds) {
queryBondInMcs.set(bond->getIdx());
}
const auto &bondRings = ri->bondRings();
for (const auto &bondRing : bondRings) {
auto ringIdx = &bondRing - &bondRings.front();
for (const auto &bondIdx : bondRing) {
if (!queryBondInMcs.test(bondIdx)) {
mcsRingIsComplete.reset(ringIdx);
break;
}
}
}
}
for (const auto &atom : mcsIdx.Atoms) {
auto queryAtomIdx = atom->getIdx();
auto numAtomRings = ri->numAtomRings(queryAtomIdx);
QueryAtom a;
const auto ai = &atom - &mcsIdx.Atoms.front();
if (Parameters.AtomTyper == MCSAtomCompareIsotopes ||
Parameters.AtomCompareParameters
.MatchIsotope) { // do '[0*]-[0*]-[13*]' for CC[13NH2]
a.setQuery(makeAtomIsotopeQuery(static_cast<int>(atom->getIsotope())));
} else {
// generate [#6] instead of C or c !
a.setQuery(makeAtomNumQuery(atom->getAtomicNum()));
// for all atomMatchSet[ai] items add atom query to template like
// [#6,#17,#9, ... ]
for (const auto &am : atomMatchSet.at(ai)) {
a.expandQuery(makeAtomNumQuery(am.second->getAtomicNum()),
Queries::COMPOSITE_OR);
if (Parameters.AtomCompareParameters.MatchChiralTag &&
(am.second->getChiralTag() == Atom::CHI_TETRAHEDRAL_CW ||
am.second->getChiralTag() == Atom::CHI_TETRAHEDRAL_CCW)) {
a.setChiralTag(am.second->getChiralTag());
}
}
}
if (needAtomRingQueries) {
const auto &ringIndicesAtomIsMemberOf = ri->atomMembers(queryAtomIdx);
auto numCompleteRings = std::count_if(
ringIndicesAtomIsMemberOf.begin(), ringIndicesAtomIsMemberOf.end(),
[&mcsRingIsComplete](const auto &ringIdx) {
return mcsRingIsComplete.test(ringIdx);
});
if (Parameters.AtomCompareParameters.RingMatchesRingOnly &&
!numCompleteRings) {
auto q = makeAtomInRingQuery();
q->setNegation(!numAtomRings);
a.expandQuery(q, Queries::COMPOSITE_AND, true);
} else if (Parameters.BondCompareParameters.MatchFusedRingsStrict &&
numAtomRings == 1 && numCompleteRings == 1) {
auto ringSize =
ri->atomRings().at(ringIndicesAtomIsMemberOf.front()).size();
auto q = new ATOM_OR_QUERY;
q->setDescription("AtomOr");
q->addChild(QueryAtom::QUERYATOM_QUERY::CHILD_TYPE(
makeAtomMinRingSizeQuery(ringSize)));
auto q2 = makeAtomInNRingsQuery(1);
q2->setNegation(true);
q->addChild(QueryAtom::QUERYATOM_QUERY::CHILD_TYPE(q2));
a.expandQuery(q, Queries::COMPOSITE_AND, true);
}
}
mol->addAtom(&a, true, false);
}
for (const auto &bond : mcsIdx.Bonds) {
QueryBond b;
const auto bi = &bond - &mcsIdx.Bonds.front();
const auto beginAtomIdx = atomIdxMap.at(bond->getBeginAtomIdx());
const auto endAtomIdx = atomIdxMap.at(bond->getEndAtomIdx());
b.setBeginAtomIdx(beginAtomIdx);
b.setEndAtomIdx(endAtomIdx);
b.setQuery(makeBondOrderEqualsQuery(bond->getBondType()));
// add OR template if need
for (const auto &bm : bondMatchSet.at(bi)) {
b.expandQuery(makeBondOrderEqualsQuery(bm.second->getBondType()),
Queries::COMPOSITE_OR);
if (Parameters.BondCompareParameters.MatchStereo &&
bm.second->getStereo() > Bond::STEREOANY) {
b.setStereo(bm.second->getStereo());
}
}
if (Parameters.BondCompareParameters.RingMatchesRingOnly ||
Parameters.BondCompareParameters.MatchFusedRingsStrict) {
const auto numBondRings = ri->numBondRings(bond->getIdx());
auto q = makeBondIsInRingQuery();
q->setNegation(!numBondRings);
b.expandQuery(q, Queries::COMPOSITE_AND, true);
}
mol->addBond(&b, false);
}
return std::make_pair(MolToSmarts(*mol, true), molSptr);
}
bool MaximumCommonSubgraph::createSeedFromMCS(size_t newQueryTarget,
Seed &newSeed) {
Seed mcs;
mcs.setStoreAllDegenerateMCS(Parameters.StoreAll);
mcs.ExcludedBonds.resize(McsIdx.QueryMolecule->getNumBonds(), false);
std::vector<unsigned int> mcsAtomIdxMap(McsIdx.QueryMolecule->getNumAtoms());
for (const auto &atom : McsIdx.Atoms) {
mcsAtomIdxMap[atom->getIdx()] = mcs.addAtom(atom);
}
for (const auto &bond : McsIdx.Bonds) {
mcs.addBond(bond);
}
const Target &newQuery = McsIdx.Targets.at(newQueryTarget);
match_V_t match;
bool target_matched = SubstructMatchCustomTable(
newQuery.Topology, *newQuery.Molecule, mcs.Topology,
*McsIdx.QueryMolecule, newQuery.AtomMatchTable, newQuery.BondMatchTable,
&Parameters, &match);
if (!target_matched) {
return false;
}
AtomMatchSet atomMatchResult(mcs.getNumAtoms());
newSeed.ExcludedBonds.resize(newQuery.Molecule->getNumBonds(), false);
for (const auto &m : match) {
unsigned int ai = m.first; // SeedAtomIdx in mcs seed
atomMatchResult[ai].QueryAtomIdx = mcs.Topology[m.first];
atomMatchResult[ai].TargetAtomIdx = newQuery.Topology[m.second];
const auto ta =
newQuery.Molecule->getAtomWithIdx(newQuery.Topology[m.second]);
newSeed.addAtom(ta);
}
for (const auto &bond : McsIdx.Bonds) {
unsigned int i = mcsAtomIdxMap.at(bond->getBeginAtomIdx());
unsigned int j = mcsAtomIdxMap.at(bond->getEndAtomIdx());
unsigned int ti = atomMatchResult.at(i).TargetAtomIdx;
unsigned int tj = atomMatchResult.at(j).TargetAtomIdx;
const auto tb = newQuery.Molecule->getBondBetweenAtoms(ti, tj);
CHECK_INVARIANT(tb, "tb most not be NULL");
newSeed.addBond(tb);
}
newSeed.computeRemainingSize(*newQuery.Molecule);
return true;
}
MCSResult MaximumCommonSubgraph::find(const std::vector<ROMOL_SPTR> &src_mols) {
clear();
MCSResult res;
if (src_mols.size() < 2) {
throw std::runtime_error(
"FMCS. Invalid argument. mols.size() must be at least 2");
}
if (Parameters.Threshold > 1.0) {
throw std::runtime_error(
"FMCS. Invalid argument. Parameter Threshold must be 1.0 or "
"less.");
}
// minimal required number of matched targets:
// at least one target, max all targets
ThresholdCount = static_cast<unsigned int>(std::min(
static_cast<int>(src_mols.size()) - 1,
std::max(1, static_cast<int>(ceil(static_cast<double>(src_mols.size()) *
Parameters.Threshold)) -
1)));
// AtomCompareParameters.CompleteRingsOnly implies
// BondCompareParameters.CompleteRingsOnly
if (Parameters.AtomCompareParameters.CompleteRingsOnly) {
Parameters.BondCompareParameters.CompleteRingsOnly = true;
}
// Selecting CompleteRingsOnly option also enables
// --ring-matches-ring-only. ring--ring and chain bonds only match chain
// bonds.
if (Parameters.BondCompareParameters.CompleteRingsOnly) {
Parameters.BondCompareParameters.RingMatchesRingOnly = true;
}
if (Parameters.AtomCompareParameters.CompleteRingsOnly) {
Parameters.AtomCompareParameters.RingMatchesRingOnly = true;
}
unsigned int i = 0;
boost::dynamic_bitset<> faked_ring_info(src_mols.size());
for (const auto &src_mol : src_mols) {
Molecules.push_back(src_mol.get());
if (!Molecules.back()->getRingInfo()->isInitialized()) {
Molecules.back()->getRingInfo()->initialize(); // but do not fill out !!!
faked_ring_info.set(i);
}
++i;
}
// sort source set of molecules by their 'size' and assume the smallest
// molecule as a query
std::stable_sort(Molecules.begin(), Molecules.end(), molPtr_NumBondLess);
size_t startIdx = 0;
size_t endIdx = Molecules.size() - ThresholdCount;
while (startIdx < endIdx && !Molecules.at(startIdx)->getNumAtoms()) {
++startIdx;
}
bool areSeedsEmpty = false;
for (size_t i = startIdx; i < endIdx && !areSeedsEmpty && !res.Canceled;
++i) {
init(startIdx);
if (Targets.empty()) {
break;
}
MCSFinalMatchCheckFunction tff = Parameters.FinalMatchChecker;
// skip final match check for initial seed to allow future growing
Parameters.FinalMatchChecker = nullptr;
makeInitialSeeds();
Parameters.FinalMatchChecker = tff; // restore final functor
if (Parameters.Verbose) {
std::cout << "Query " << MolToSmiles(*QueryMolecule) << " "
<< QueryMolecule->getNumAtoms() << "("
<< QueryMoleculeMatchedAtoms << ") atoms, "
<< QueryMolecule->getNumBonds() << "("
<< QueryMoleculeMatchedBonds << ") bonds\n";
}
areSeedsEmpty = Seeds.empty();
res.Canceled = !(areSeedsEmpty || growSeeds());
// verify what MCS is equal to one of initial seed for chirality match
if (getMaxNumberBonds() == 0) {
McsIdx = MCS(); // clear
makeInitialSeeds(); // check all possible initial seeds
if (!areSeedsEmpty) {
const Seed &fs = Seeds.front();
if ((1 == getMaxNumberBonds() ||
!(Parameters.BondCompareParameters.CompleteRingsOnly &&
fs.MoleculeFragment.Bonds.size() == 1 &&
queryIsBondInRing(fs.MoleculeFragment.Bonds.front()))) &&
checkIfShouldAcceptMCS(fs.MoleculeFragment, *QueryMolecule, Targets,
Parameters)) {
McsIdx.QueryMolecule = QueryMolecule;
McsIdx.Targets = Targets;
McsIdx.Atoms = fs.MoleculeFragment.Atoms;
McsIdx.Bonds = fs.MoleculeFragment.Bonds;
}
}
if (!McsIdx.QueryMolecule && QueryMoleculeSingleMatchedAtom) {
McsIdx.QueryMolecule = QueryMolecule;
McsIdx.Targets = Targets;
McsIdx.Atoms =
std::vector<const Atom *>{QueryMoleculeSingleMatchedAtom};
McsIdx.Bonds = std::vector<const Bond *>();
}
} else if (i + 1 < endIdx) {
Seed seed;
if (createSeedFromMCS(i, seed)) { // MCS is matched with new query
Seeds.push_back(seed);
}
std::swap(
Molecules.at(startIdx),
Molecules.at(i + 1)); // change query molecule for threshold < 1.
}
}
res.NumAtoms = getMaxNumberAtoms();
if (!res.NumAtoms && QueryMoleculeSingleMatchedAtom) {
res.NumAtoms = 1;
}
res.NumBonds = getMaxNumberBonds();
if (res.NumBonds > 0 || QueryMoleculeSingleMatchedAtom) {
if (!Parameters.StoreAll) {
auto smartsQueryMolPair = generateResultSMARTSAndQueryMol(McsIdx);
res.SmartsString = std::move(smartsQueryMolPair.first);
res.QueryMol = std::move(smartsQueryMolPair.second);
} else {
std::transform(DegenerateMcsMap.begin(), DegenerateMcsMap.end(),
std::inserter(res.DegenerateSmartsQueryMolDict,
res.DegenerateSmartsQueryMolDict.end()),
[this](const auto &pair) {
return generateResultSMARTSAndQueryMol(pair.second);
});
}
}
#ifdef VERBOSE_STATISTICS_ON
if (Parameters.Verbose && res.NumAtoms > 0) {
for (const auto &tag : Targets) {
unsigned int itarget = &tag - &Targets.front();
MatchVectType match;
bool target_matched = SubstructMatch(*tag.Molecule, *res.QueryMol, match);
if (!target_matched) {
std::cout << "Target " << itarget + 1
<< (target_matched ? " matched " : " MISMATCHED ")
<< MolToSmiles(*tag.Molecule) << "\n";
}
}
std::cout << "STATISTICS:\n";
std::cout << "Total Growing Steps = " << VerboseStatistics.TotalSteps
<< ", MCS found on " << VerboseStatistics.MCSFoundStep << " step";
if (VerboseStatistics.MCSFoundTime - To > 0) {
printf(", for %.4lf seconds\n",
double(VerboseStatistics.MCSFoundTime - To) / 1000000.);
} else {
std::cout << ", for less than 1 second\n";
}
std::cout << "Initial Seeds = " << VerboseStatistics.InitialSeed
<< ", Mismatched " << VerboseStatistics.MismatchedInitialSeed
<< "\n";
std::cout << "Inspected Seeds = " << VerboseStatistics.Seed << "\n";
std::cout << "Rejected by BestSize = "
<< VerboseStatistics.RemainingSizeRejected << "\n";
std::cout << "IndividualBondExcluded = "
<< VerboseStatistics.IndividualBondExcluded << "\n";
#ifdef EXCLUDE_WRONG_COMPOSITION
std::cout << "Rejected by WrongComposition = "
<< VerboseStatistics.WrongCompositionRejected << " [ "
<< VerboseStatistics.WrongCompositionDetected << " Detected ]\n";
#endif
std::cout << "MatchCheck Seeds = " << VerboseStatistics.SeedCheck
<< "\n";
std::cout //<< "\n"
<< " MatchCalls = " << VerboseStatistics.MatchCall << "\n"
<< " MatchFound = " << VerboseStatistics.MatchCallTrue << "\n";
std::cout << " fastMatchCalls = " << VerboseStatistics.FastMatchCall << "\n"
<< " fastMatchFound = " << VerboseStatistics.FastMatchCallTrue
<< "\n";
std::cout << " slowMatchCalls = "
<< VerboseStatistics.MatchCall -
VerboseStatistics.FastMatchCallTrue
<< "\n"
<< " slowMatchFound = " << VerboseStatistics.SlowMatchCallTrue
<< "\n";
#ifdef VERBOSE_STATISTICS_FASTCALLS_ON
std::cout << "AtomFunctorCalls = " << VerboseStatistics.AtomFunctorCalls
<< "\n";
std::cout << "BondCompareCalls = " << VerboseStatistics.BondCompareCalls
<< "\n";
#endif
std::cout << " DupCacheFound = " << VerboseStatistics.DupCacheFound
<< " " << VerboseStatistics.DupCacheFoundMatch << " matched, "
<< VerboseStatistics.DupCacheFound -
VerboseStatistics.DupCacheFoundMatch
<< " mismatched\n";
#ifdef FAST_SUBSTRUCT_CACHE
std::cout << "HashCache size = " << HashCache.keyssize() << " keys\n";
std::cout << "HashCache size = " << HashCache.fullsize() << " entries\n";
std::cout << "FindHashInCache = " << VerboseStatistics.FindHashInCache
<< "\n";
std::cout << "HashFoundInCache= " << VerboseStatistics.HashKeyFoundInCache
<< "\n";
std::cout << "ExactMatchCalls = " << VerboseStatistics.ExactMatchCall
<< "\n"
<< "ExactMatchFound = " << VerboseStatistics.ExactMatchCallTrue
<< "\n";
#endif
}
#endif
auto pos = faked_ring_info.find_first();
while (pos != boost::dynamic_bitset<>::npos) {
src_mols[pos]->getRingInfo()->reset();
pos = faked_ring_info.find_next(pos);
}
clear();
return res;
}
bool MaximumCommonSubgraph::checkIfMatchAndAppend(Seed &seed) {
#ifdef VERBOSE_STATISTICS_ON
++VerboseStatistics.SeedCheck;
#endif
#ifdef FAST_SUBSTRUCT_CACHE
SubstructureCache::HashKey cacheKey;
SubstructureCache::TIndexEntry *cacheEntry = nullptr;
#endif
bool foundInCache = false;
bool foundInDupCache = false;
{
#ifdef DUP_SUBSTRUCT_CACHE
if (DuplicateCache.find(seed.DupCacheKey, foundInCache)) {
// duplicate found. skip match() but store both seeds, because they will grow by
// different paths !!!
#ifdef VERBOSE_STATISTICS_ON
VerboseStatistics.DupCacheFound++;
VerboseStatistics.DupCacheFoundMatch += foundInCache ? 1 : 0;
#endif
if (!foundInCache) { // mismatched !!!
return false;
}
}
foundInDupCache = foundInCache;
#endif
#ifdef FAST_SUBSTRUCT_CACHE
if (!foundInCache) {
#ifdef VERBOSE_STATISTICS_ON
++VerboseStatistics.FindHashInCache;
#endif
cacheEntry =
HashCache.find(seed, QueryAtomLabels, QueryBondLabels, cacheKey);
if (cacheEntry) { // possibly found. check for hash collision
#ifdef VERBOSE_STATISTICS_ON
++VerboseStatistics.HashKeyFoundInCache;
#endif
// check hash collisions (time +3%):
for (const auto &g : *cacheEntry) {
if (g.m_vertices.size() != seed.getNumAtoms() ||
g.m_edges.size() != seed.getNumBonds()) {
continue;
}
#ifdef VERBOSE_STATISTICS_ON
++VerboseStatistics.ExactMatchCall;
#endif
// EXACT MATCH
foundInCache = SubstructMatchCustomTable(
g, *QueryMolecule, seed.Topology, *QueryMolecule,
QueryAtomMatchTable, QueryBondMatchTable, &Parameters);
#ifdef VERBOSE_STATISTICS_ON
if (foundInCache) {
++VerboseStatistics.ExactMatchCallTrue;
} else {
break;
}
#endif
}
}
}
#endif
}
bool found = foundInCache;
if (!found) {
found = match(seed);
}
Seed *newSeed = nullptr;
{
if (found) { // Store new generated seed, if found in cache or in
// all(- threshold) targets
newSeed = &Seeds.add(seed);
newSeed->CopyComplete = false;
#ifdef DUP_SUBSTRUCT_CACHE
if (!foundInDupCache &&
seed.getNumBonds() >= 3) { // only seed with a ring
// can be duplicated -
// do not store very
// small seed in cache
DuplicateCache.add(seed.DupCacheKey, true);
}
#endif
#ifdef FAST_SUBSTRUCT_CACHE
if (!foundInCache) {
HashCache.add(seed, cacheKey, cacheEntry);
}
#endif
} else {
#ifdef DUP_SUBSTRUCT_CACHE
if (seed.getNumBonds() > 3) {
DuplicateCache.add(seed.DupCacheKey,
false); // opt. cache mismatched duplicates too
}
#endif
}
}
if (newSeed) {
*newSeed = seed; // non-blocking copy for MULTI_THREAD and best CPU
// utilization
}
return found; // new matched seed has been actually added
}
bool MaximumCommonSubgraph::match(Seed &seed) {
unsigned int max_miss = Targets.size() - ThresholdCount;
unsigned int missing = 0;
unsigned int passed = 0;
for (const auto &tag : Targets) {
unsigned int itarget = &tag - &Targets.front();
#ifdef VERBOSE_STATISTICS_ON
{ ++VerboseStatistics.MatchCall; }
#endif
bool target_matched = false;
if (!seed.MatchResult.empty() && !seed.MatchResult.at(itarget).empty()) {
target_matched = matchIncrementalFast(seed, itarget);
}
if (!target_matched) { // slow full match
match_V_t match; // THERE IS NO Bonds match INFO !!!!
target_matched = SubstructMatchCustomTable(
tag.Topology, *tag.Molecule, seed.Topology, *QueryMolecule,
tag.AtomMatchTable, tag.BondMatchTable, &Parameters, &match);
// save current match info
if (target_matched) {
if (seed.MatchResult.empty()) {
seed.MatchResult.resize(Targets.size());
}
seed.MatchResult[itarget].init(seed, match, *QueryMolecule, tag);
} else if (!seed.MatchResult.empty()) {
seed.MatchResult[itarget].clear(); //.Empty = true; // == fast clear();
}
#ifdef VERBOSE_STATISTICS_ON
if (target_matched) {
++VerboseStatistics.SlowMatchCallTrue;
}
#endif
}
if (target_matched) {
if (++passed >= ThresholdCount) { // it's enough
break;
}
} else { // mismatched
if (++missing > max_miss) {
break;
}
}
}
if (missing <= max_miss) {
#ifdef VERBOSE_STATISTICS_ON
++VerboseStatistics.MatchCallTrue;
#endif
return true;
}
return false;
}
// call it for each target, if failed perform full match check
bool MaximumCommonSubgraph::matchIncrementalFast(Seed &seed,
unsigned int itarget) {
// use and update results of previous match stored in the seed
#ifdef VERBOSE_STATISTICS_ON
{ ++VerboseStatistics.FastMatchCall; }
#endif
const auto &target = Targets.at(itarget);
auto &match = seed.MatchResult.at(itarget);
if (match.empty()) {
return false;
}
/*
// CHIRALITY: FinalMatchCheck:
if(Parameters.AtomCompareParameters.MatchChiralTag ||
Parameters.FinalMatchChecker) { // TEMP
match.clear();
return false;
}
*/
bool matched = false;
for (unsigned int newBondSeedIdx = match.MatchedBondSize;
newBondSeedIdx < seed.getNumBonds(); newBondSeedIdx++) {
matched = false;
bool atomAdded = false;
const auto newBond = seed.MoleculeFragment.Bonds.at(newBondSeedIdx);
unsigned int newBondQueryIdx = newBond->getIdx();
// seed's index of atom from which new bond was added
unsigned int newBondSourceAtomSeedIdx;
// seed's index of atom on other end of the bond
unsigned int newBondOtherAtomSeedIdx;
unsigned int i =
seed.MoleculeFragment.SeedAtomIdxMap.at(newBond->getBeginAtomIdx());
unsigned int j =
seed.MoleculeFragment.SeedAtomIdxMap.at(newBond->getEndAtomIdx());
if (i >= match.MatchedAtomSize) {
// this is new atom in the seed
newBondSourceAtomSeedIdx = j;
newBondOtherAtomSeedIdx = i;
} else {
newBondSourceAtomSeedIdx = i;
newBondOtherAtomSeedIdx = j;
}
unsigned int newBondOtherAtomQueryIdx =
seed.MoleculeFragment.Atoms.at(newBondOtherAtomSeedIdx)->getIdx();
unsigned int newBondSourceAtomQueryIdx =
seed.MoleculeFragment.Atoms.at(newBondSourceAtomSeedIdx)->getIdx();
// matched to newBondSourceAtomSeedIdx
unsigned int newBondSourceAtomTargetIdx =
match.TargetAtomIdx.at(newBondSourceAtomQueryIdx);
const Bond *tb = nullptr;
unsigned int newBondOtherAtomTargetIdx = NotSet;
if (newBondOtherAtomSeedIdx < match.MatchedAtomSize) {
// new bond between old atoms - both are
// matched to exact atoms in the target
newBondOtherAtomTargetIdx =
match.TargetAtomIdx.at(newBondOtherAtomQueryIdx);
// target bond between Source and Other atom
tb = target.Molecule->getBondBetweenAtoms(newBondSourceAtomTargetIdx,
newBondOtherAtomTargetIdx);
if (tb) {
// bond exists, check match with query molecule
unsigned int tbi = tb->getIdx();
unsigned int qbi =
seed.MoleculeFragment.Bonds.at(newBondSeedIdx)->getIdx();
if (!match.VisitedTargetBonds.test(tbi)) {
// false if target bond is already matched
matched = target.BondMatchTable.at(qbi, tbi);
}
}
} else {
// enumerate all bonds from source atom in the target
const auto atom =
target.Molecule->getAtomWithIdx(newBondSourceAtomTargetIdx);
for (const auto &nbri :
boost::make_iterator_range(target.Molecule->getAtomBonds(atom))) {
tb = (*target.Molecule)[nbri];
if (match.VisitedTargetBonds.test(tb->getIdx())) {
continue;
}
newBondOtherAtomTargetIdx = tb->getBeginAtomIdx();
if (newBondSourceAtomTargetIdx == newBondOtherAtomTargetIdx) {
newBondOtherAtomTargetIdx = tb->getEndAtomIdx();
}
if (match.VisitedTargetAtoms.test(newBondOtherAtomTargetIdx)) {
continue;
}
// check OtherAtom and bond
matched = target.AtomMatchTable.at(newBondOtherAtomQueryIdx,
newBondOtherAtomTargetIdx) &&
target.BondMatchTable.at(
seed.MoleculeFragment.Bonds.at(newBondSeedIdx)->getIdx(),
tb->getIdx());
if (matched) {
atomAdded = true;
break;
}
}
}
if (matched) { // update match history
if (atomAdded) { // new atom has been added
match.MatchedAtomSize++;
match.TargetAtomIdx[newBondOtherAtomQueryIdx] =
newBondOtherAtomTargetIdx;
match.VisitedTargetAtoms.set(newBondOtherAtomTargetIdx);
}
match.MatchedBondSize++;
match.TargetBondIdx[newBondQueryIdx] = tb->getIdx();
match.VisitedTargetBonds.set(tb->getIdx());
} else {
match.clear();
return false;
}
}
if (match.MatchedAtomSize != seed.getNumAtoms() ||
match.MatchedBondSize !=
seed.getNumBonds()) { // number of unique items !!!
match.clear();
return false;
}
// CHIRALITY: FinalMatchCheck
if (matched && Parameters.FinalMatchChecker) {
std::vector<std::uint32_t> c1;
c1.reserve(seed.getNumAtoms());
std::vector<std::uint32_t> c2;
c2.reserve(target.Molecule->getNumAtoms());
for (unsigned int si = 0; si < seed.getNumAtoms(); ++si) {
// index in the seed topology
c1.push_back(si);
c2.push_back(match.TargetAtomIdx.at(seed.Topology[si]));
}
matched = Parameters.FinalMatchChecker(c1.data(), c2.data(), *QueryMolecule,
seed.Topology, *target.Molecule,
target.Topology,
&Parameters); // check CHIRALITY
if (!matched) {
match.clear();
}
}
#ifdef VERBOSE_STATISTICS_ON
if (matched) {
#ifdef MULTI_THREAD
Guard statlock(StatisticsMutex);
#endif
++VerboseStatistics.FastMatchCallTrue;
}
#endif
return matched;
}
} // namespace FMCS
} // namespace RDKit
|