// // Copyright (C) 2018 Boran Adas, Google Summer of Code // // @@ 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 #ifndef RD_MORGANGEN_H_2018_07 #define RD_MORGANGEN_H_2018_07 #include #include namespace RDKit { namespace MorganFingerprint { /** \brief Default atom invariants generator for Morgan fingerprint, generates ECFP-type invariants */ class RDKIT_FINGERPRINTS_EXPORT MorganAtomInvGenerator : public AtomInvariantsGenerator { const bool df_includeRingMembership; public: /** \brief Construct a new MorganAtomInvGenerator object \param includeRingMembership : if set, whether or not the atom is in a ring will be used in the invariant list. */ MorganAtomInvGenerator(const bool includeRingMembership = true); std::vector *getAtomInvariants( const ROMol &mol) const override; std::string infoString() const override; MorganAtomInvGenerator *clone() const override; }; /** \brief Alternative atom invariants generator for Morgan fingerprint, generate FCFP-type invariants */ class RDKIT_FINGERPRINTS_EXPORT MorganFeatureAtomInvGenerator : public AtomInvariantsGenerator { std::vector *dp_patterns; public: /** \brief Construct a new MorganFeatureAtomInvGenerator object \param patterns : if provided should contain the queries used to assign atom-types. if not provided, feature definitions adapted from reference: Gobbi and Poppinger, Biotech. Bioeng. _61_ 47-54 (1998) will be used for Donor, Acceptor, Aromatic, Halogen, Basic, Acidic. */ MorganFeatureAtomInvGenerator(std::vector *patterns = nullptr); std::vector *getAtomInvariants( const ROMol &mol) const override; std::string infoString() const override; MorganFeatureAtomInvGenerator *clone() const override; }; /** \brief Bond invariants generator for Morgan fingerprint */ class RDKIT_FINGERPRINTS_EXPORT MorganBondInvGenerator : public BondInvariantsGenerator { const bool df_useBondTypes; const bool df_useChirality; public: /** \brief Construct a new MorganBondInvGenerator object \param useBondTypes : if set, bond types will be included as a part of the bond invariants \param useChirality : if set, chirality information will be included as a part of the bond invariants */ MorganBondInvGenerator(const bool useBondTypes = true, const bool useChirality = false); std::vector *getBondInvariants( const ROMol &mol) const override; std::string infoString() const override; MorganBondInvGenerator *clone() const override; ~MorganBondInvGenerator() override = default; }; /** \brief Class for holding Morgan fingerprint specific arguments */ template class RDKIT_FINGERPRINTS_EXPORT MorganArguments : public FingerprintArguments { public: const bool df_includeChirality; const bool df_onlyNonzeroInvariants; const unsigned int d_radius; OutputType getResultSize() const override; std::string infoString() const override; /** \brief Construct a new MorganArguments object \param radius the number of iterations to grow the fingerprint \param countSimulation if set, use count simulation while generating the fingerprint \param includeChirality if set, chirality information will be added to the generated bit id, independently from bond invariants \param onlyNonzeroInvariants if set, bits will only be set from atoms that have a nonzero invariant \param countBounds boundaries for count simulation, corresponding bit will be set if the count is higher than the number provided for that spot \param fpSize size of the generated fingerprint, does not affect the sparse versions */ MorganArguments(const unsigned int radius, const bool countSimulation = false, const bool includeChirality = false, const bool onlyNonzeroInvariants = false, const std::vector countBounds = {1, 2, 4, 8}, const std::uint32_t fpSize = 2048); }; /** \brief Class for holding the bit-id created from Morgan fingerprint environments and the additional data necessary extra outputs */ template class RDKIT_FINGERPRINTS_EXPORT MorganAtomEnv : public AtomEnvironment { const OutputType d_code; const unsigned int d_atomId; const unsigned int d_layer; public: OutputType getBitId(FingerprintArguments *arguments, const std::vector *atomInvariants, const std::vector *bondInvariants, const AdditionalOutput *additionalOutput, const bool hashResults = false, const std::uint64_t fpSize = 0) const override; /** \brief Construct a new MorganAtomEnv object \param code bit id generated from this environment \param atomId atom id of the atom at the center of this environment \param layer radius of this environment */ MorganAtomEnv(const std::uint32_t code, const unsigned int atomId, const unsigned int layer); }; /** \brief Class that generates atom environments for Morgan fingerprint */ template class RDKIT_FINGERPRINTS_EXPORT MorganEnvGenerator : public AtomEnvironmentGenerator { public: std::vector *> getEnvironments( const ROMol &mol, FingerprintArguments *arguments, const std::vector *fromAtoms, const std::vector *ignoreAtoms, const int confId, const AdditionalOutput *additionalOutput, const std::vector *atomInvariants, const std::vector *bondInvariants, const bool hashResults = false) const override; std::string infoString() const override; }; /** \brief Get a fingerprint generator for Morgan fingerprint \tparam OutputType determines the size of the bitIds and the result, can be 32 or 64 bit unsigned integer \param radius the number of iterations to grow the fingerprint \param countSimulation if set, use count simulation while generating the fingerprint \param includeChirality if set, chirality information will be added to the generated bit id, independently from bond invariants \param onlyNonzeroInvariants if set, bits will only be set from atoms that have a nonzero invariant \param countBounds boundaries for count simulation, corresponding bit will be set if the count is higher than the number provided for that spot \param fpSize size of the generated fingerprint, does not affect the sparse versions \param countSimulation if set, use count simulation while generating the fingerprint \param includeChirality sets includeChirality flag for both MorganArguments and the default bond generator MorganBondInvGenerator \param useBondTypes if set, bond types will be included as a part of the default bond invariants \param onlyNonzeroInvariants if set, bits will only be set from atoms that have a nonzero invariant \param atomInvariantsGenerator custom atom invariants generator to use \param bondInvariantsGenerator custom bond invariants generator to use \param ownsAtomInvGen if set atom invariants generator is destroyed with the fingerprint generator \param ownsBondInvGen if set bond invariants generator is destroyed with the fingerprint generator \return FingerprintGenerator* that generates Morgan fingerprints This generator supports the following \c AdditionalOutput types: - \c atomToBits : which bits each atom is the central atom for - \c atomCounts : how many bits each atom sets - \c bitInfoMap : map from bitId to (atomId, radius) pairs */ template RDKIT_FINGERPRINTS_EXPORT FingerprintGenerator *getMorganGenerator( const unsigned int radius, const bool countSimulation = false, const bool includeChirality = false, const bool useBondTypes = true, const bool onlyNonzeroInvariants = false, AtomInvariantsGenerator *atomInvariantsGenerator = nullptr, BondInvariantsGenerator *bondInvariantsGenerator = nullptr, const std::uint32_t fpSize = 2048, const std::vector countBounds = {1, 2, 4, 8}, const bool ownsAtomInvGen = false, const bool ownsBondInvGen = false); } // namespace MorganFingerprint } // namespace RDKit #endif