RDKit
Open-source cheminformatics and machine learning.
Seed.h
Go to the documentation of this file.
1//
2// Copyright (C) 2014 Novartis Institutes for BioMedical Research
3//
4// @@ All Rights Reserved @@
5// This file is part of the RDKit.
6// The contents are covered by the terms of the BSD license
7// which is included in the file license.txt, found at the root
8// of the RDKit source tree.
9//
10#include <RDGeneral/export.h>
11#pragma once
12#include <map>
13#include "../RDKitBase.h"
14#include "DebugTrace.h" // algorithm optimisation definitions
15#include "Graph.h"
16#include "DuplicatedSeedCache.h"
18
19namespace RDKit {
20namespace FMCS {
21class MaximumCommonSubgraph;
22struct TargetMatch;
23
25 MolFragment { // Reference to a fragment of source molecule
26 std::vector<const Atom*> Atoms;
27 std::vector<const Bond*> Bonds;
28 std::vector<unsigned> AtomsIdx;
29 std::vector<unsigned> BondsIdx; // need for results and size() only !
30 std::map<unsigned, unsigned> SeedAtomIdxMap; // Full Query Molecule to Seed
31 // indices backward conversion
32 // map
33};
34
36 unsigned SourceAtomIdx{0}; // index in the seed. Atom is already in the seed
37 unsigned BondIdx{0}; // index in qmol of new bond scheduled to be added into
38 // seed. This is outgoing bond from SourceAtomIdx
39 unsigned NewAtomIdx{0}; // index in qmol of new atom scheduled to be added
40 // into seed. Another end of new bond
41 const Atom* NewAtom{nullptr}; // pointer to qmol's new atom scheduled to be
42 // added into seed. Another end of new bond
43 unsigned EndAtomIdx{0}; // index in the seed. RING. "New" Atom on the another
44 // end of new bond is already exists in the seed.
45
47
48 {}
49
50 NewBond(unsigned from_atom, unsigned bond_idx, unsigned new_atom,
51 unsigned to_atom, const Atom* a)
52 : SourceAtomIdx(from_atom),
53 BondIdx(bond_idx),
54 NewAtomIdx(new_atom),
55 NewAtom(a),
56 EndAtomIdx(to_atom) {}
57};
58
60 private:
61 mutable std::vector<NewBond> NewBonds; // for multistage growing. all
62 // directly connected outgoing bonds
63 public:
64 bool CopyComplete{false}; // this seed has been completely copied into list.
65 // postponed non-locked copy for MULTI_THREAD
66 mutable unsigned GrowingStage{0}; // 0 new seed; -1 finished; n>0 in
67 // progress, exact stage of growing for SDF
68 MolFragment MoleculeFragment; // Reference to a fragment of source molecule
69 Graph Topology; // seed topology with references to source molecule
70
71 std::vector<bool> ExcludedBonds;
72 unsigned LastAddedAtomsBeginIdx{0}; // in this subgraph for improving
73 // performance of future growing
74 unsigned LastAddedBondsBeginIdx{0}; // in this subgraph for DEBUG ONLY
75 unsigned RemainingBonds{0};
76 unsigned RemainingAtoms{0};
77#ifdef DUP_SUBSTRUCT_CACHE
79#endif
80 std::vector<TargetMatch> MatchResult; // for each target
81 public:
83
84 {}
85
86 void setMoleculeFragment(const Seed& src) {
87 MoleculeFragment = src.MoleculeFragment;
88 }
89 Seed& operator=(const Seed& src) {
90 NewBonds = src.NewBonds;
91 GrowingStage = src.GrowingStage;
92 MoleculeFragment = src.MoleculeFragment;
93 Topology = src.Topology;
94 ExcludedBonds = src.ExcludedBonds;
95 LastAddedAtomsBeginIdx = src.LastAddedAtomsBeginIdx;
96 LastAddedBondsBeginIdx = src.LastAddedBondsBeginIdx;
97 RemainingBonds = src.RemainingBonds;
98 RemainingAtoms = src.RemainingAtoms;
99#ifdef DUP_SUBSTRUCT_CACHE
100 DupCacheKey = src.DupCacheKey;
101#endif
102 MatchResult = src.MatchResult;
103 CopyComplete = true; // LAST
104 return *this;
105 }
106 void createFromParent(const Seed* parent) {
107 MoleculeFragment = parent->MoleculeFragment;
108 Topology = parent->Topology;
109 ExcludedBonds = parent->ExcludedBonds;
110 RemainingBonds = parent->RemainingBonds;
111 RemainingAtoms = parent->RemainingAtoms;
112#ifdef DUP_SUBSTRUCT_CACHE
113 DupCacheKey = parent->DupCacheKey;
114#endif
115 LastAddedAtomsBeginIdx = getNumAtoms(); // previous size
116 LastAddedBondsBeginIdx = getNumBonds(); // previous size
117 GrowingStage = 0;
118 }
119
120 unsigned getNumAtoms() const { return MoleculeFragment.AtomsIdx.size(); }
121 unsigned getNumBonds() const { return MoleculeFragment.BondsIdx.size(); }
122
123 void grow(MaximumCommonSubgraph& mcs) const;
124 bool canGrowBiggerThan(unsigned maxBonds,
125 unsigned maxAtoms) const { // prune()
126 return RemainingBonds + getNumBonds() > maxBonds ||
127 (RemainingBonds + getNumBonds() == maxBonds &&
128 RemainingAtoms + getNumAtoms() > maxAtoms);
129 }
130 void computeRemainingSize(const ROMol& qmol);
131
132 unsigned addAtom(const Atom* atom);
133 unsigned addBond(const Bond* bond);
134 void fillNewBonds(const ROMol& qmol);
135};
136} // namespace FMCS
137} // namespace RDKit
The class for representing atoms.
Definition: Atom.h:68
class for representing a bond
Definition: Bond.h:47
unsigned getNumBonds() const
Definition: Seed.h:121
bool canGrowBiggerThan(unsigned maxBonds, unsigned maxAtoms) const
Definition: Seed.h:124
void createFromParent(const Seed *parent)
Definition: Seed.h:106
unsigned LastAddedBondsBeginIdx
Definition: Seed.h:74
void grow(MaximumCommonSubgraph &mcs) const
unsigned GrowingStage
Definition: Seed.h:66
unsigned getNumAtoms() const
Definition: Seed.h:120
unsigned RemainingBonds
Definition: Seed.h:75
MolFragment MoleculeFragment
Definition: Seed.h:68
DuplicatedSeedCache::TKey DupCacheKey
Definition: Seed.h:78
std::vector< TargetMatch > MatchResult
Definition: Seed.h:80
unsigned RemainingAtoms
Definition: Seed.h:76
unsigned addAtom(const Atom *atom)
void setMoleculeFragment(const Seed &src)
Definition: Seed.h:86
void fillNewBonds(const ROMol &qmol)
std::vector< bool > ExcludedBonds
Definition: Seed.h:71
Seed & operator=(const Seed &src)
Definition: Seed.h:89
unsigned LastAddedAtomsBeginIdx
Definition: Seed.h:72
void computeRemainingSize(const ROMol &qmol)
unsigned addBond(const Bond *bond)
Graph Topology
Definition: Seed.h:69
#define RDKIT_FMCS_EXPORT
Definition: export.h:153
Std stuff.
Definition: Abbreviations.h:19
std::vector< unsigned > AtomsIdx
Definition: Seed.h:28
std::vector< const Atom * > Atoms
Definition: Seed.h:26
std::vector< const Bond * > Bonds
Definition: Seed.h:27
std::vector< unsigned > BondsIdx
Definition: Seed.h:29
std::map< unsigned, unsigned > SeedAtomIdxMap
Definition: Seed.h:30
NewBond(unsigned from_atom, unsigned bond_idx, unsigned new_atom, unsigned to_atom, const Atom *a)
Definition: Seed.h:50