EmpiricalFormula.h

Go to the documentation of this file.

// Copyright (c) 2002-present, OpenMS Inc. -- EKU Tuebingen, ETH Zurich, and FU Berlin
// SPDX-License-Identifier: BSD-3-Clause
//
// --------------------------------------------------------------------------
// $Maintainer: Chris Bielow, Ahmed Khalil $
// $Authors: Andreas Bertsch, Chris Bielow $
// --------------------------------------------------------------------------
//
#pragma once
 
#include <iosfwd>
#include <algorithm>
#include <map>
#include <set>
#include <string>
#include <functional>
#include <vector>
 
#include <OpenMS/CONCEPT/Types.h>
#include <OpenMS/CONCEPT/HashUtils.h>
#include <OpenMS/CHEMISTRY/Element.h>
 
namespace OpenMS
{
  class ElementDB;
  class IsotopeDistribution;
  class IsotopePatternGenerator;
  class CoarseIsotopePatternGenerator;
 
  class OPENMS_DLLAPI EmpiricalFormula
  {
 
protected:
    struct ElementLess_
    {
      bool operator()(const Element* a, const Element* b) const
      {
        if (a == nullptr || b == nullptr) return std::less<const Element*>()(a, b);
        if (a->getAtomicNumber() != b->getAtomicNumber()) return a->getAtomicNumber() < b->getAtomicNumber();
        const std::string& symbol_a = a->getSymbol();
        const std::string& symbol_b = b->getSymbol();
        if (symbol_a.size() != symbol_b.size()) return symbol_a.size() < symbol_b.size();
        if (symbol_a != symbol_b) return symbol_a < symbol_b;
        return std::less<const Element*>()(a, b);
      }
    };
 
    typedef std::map<const Element*, SignedSize, ElementLess_> MapType_;
 
public:
    typedef MapType_::const_iterator ConstIterator;
    typedef MapType_::const_iterator const_iterator;
    typedef MapType_::iterator Iterator;
    typedef MapType_::iterator iterator;
 
    EmpiricalFormula();
 
    EmpiricalFormula(const EmpiricalFormula&) = default;
 
    EmpiricalFormula(EmpiricalFormula&&) = default;
 
    explicit EmpiricalFormula(const std::string& rhs);
 
    EmpiricalFormula(SignedSize number, const Element* element, SignedSize charge = 0);
 
    virtual ~EmpiricalFormula();
 
    static EmpiricalFormula fromString(const std::string& rhs)
    {
      EmpiricalFormula ef(rhs);
      return ef;
    }
 
    double getMonoWeight() const;
 
    double getLightestIsotopeWeight() const;
 
    double getAverageWeight() const;
 
    double calculateTheoreticalIsotopesNumber() const;
 
    bool estimateFromWeightAndComp(double average_weight, double C, double H, double N, double O, double S, double P);
 
    bool estimateFromMonoWeightAndComp(double mono_weight, double C, double H, double N, double O, double S, double P);
 
    bool estimateFromWeightAndCompAndS(double average_weight, UInt S, double C, double H, double N, double O, double P);
 
 
    IsotopeDistribution getIsotopeDistribution(const IsotopePatternGenerator& method) const;
 
    IsotopeDistribution getConditionalFragmentIsotopeDist(const EmpiricalFormula& precursor,
                                                          const std::set<UInt>& precursor_isotopes,
                                                          const CoarseIsotopePatternGenerator& method) const;
 
    SignedSize getNumberOf(const Element* element) const;
 
    SignedSize getNumberOfAtoms() const;
 
    Int getCharge() const;
 
    void setCharge(Int charge);
 
    EmpiricalFormula& addChargeAdduct(Int count = 1, const std::string& adduct = "H");
 
    std::string toString() const;
 
    std::map<std::string, int> toMap() const;
 
 
    EmpiricalFormula& operator=(const EmpiricalFormula&) = default;
 
    EmpiricalFormula& operator=(EmpiricalFormula&&) & = default;
 
    EmpiricalFormula& operator+=(const EmpiricalFormula& rhs);
 
    EmpiricalFormula operator*(const SignedSize& times) const;
 
    EmpiricalFormula operator+(const EmpiricalFormula& rhs) const;
 
    EmpiricalFormula& operator-=(const EmpiricalFormula& rhs);
 
    EmpiricalFormula operator-(const EmpiricalFormula& rhs) const;
 
 
    bool isEmpty() const;
 
    bool isCharged() const;
 
    bool hasElement(const Element* element) const;
 
    bool contains(const EmpiricalFormula& ef) const;
 
    bool operator==(const EmpiricalFormula& rhs) const;
 
    bool operator!=(const EmpiricalFormula& rhs) const;
 
    bool operator<(const EmpiricalFormula& rhs) const;
 
 
    friend OPENMS_DLLAPI std::ostream& operator<<(std::ostream& os, const EmpiricalFormula& formula);
 
    inline ConstIterator begin() const { return formula_.begin(); }
 
    inline ConstIterator end() const { return formula_.end(); }
 
    inline Iterator begin() { return formula_.begin(); }
 
    inline Iterator end() { return formula_.end(); }
 
    // @TODO: make these static member variables instead?
 
    static EmpiricalFormula hydrogen(int n_atoms = 1);
 
    static EmpiricalFormula water(int n_molecules = 1);
 
protected:
 
    void removeZeroedElements_();
 
    MapType_ formula_;
 
    Int charge_;
 
    Int parseFormula_(MapType_& ef, const std::string& formula) const;
 
  };
 
  OPENMS_DLLAPI std::ostream& operator<<(std::ostream& os, const EmpiricalFormula& formula);
 
} // namespace OpenMS
 
// Hash function specialization for EmpiricalFormula
// Placed in std namespace to allow use with std::unordered_map/set
namespace std
{
  template<>
  struct hash<OpenMS::EmpiricalFormula>
  {
    std::size_t operator()(const OpenMS::EmpiricalFormula& ef) const noexcept
    {
      // Collect elements with symbols for deterministic ordering
      // (map iteration order depends on pointer addresses which vary across runs)
      // Use symbols instead of atomic numbers to distinguish isotopes like (13)C vs C
      // Typical formulas have only 4-6 elements, so no need to reserve
      std::vector<std::pair<std::string, OpenMS::SignedSize>> elements;
      for (const auto& [element_ptr, count] : ef)
      {
        elements.emplace_back(element_ptr->getSymbol(), count);
      }
 
      // Sort by symbol for reproducible hash
      std::sort(elements.begin(), elements.end());
 
      // Hash in sorted order
      std::size_t seed = 0;
      for (const auto& [symbol, count] : elements)
      {
        OpenMS::hash_combine(seed, OpenMS::fnv1a_hash_string(symbol));
        OpenMS::hash_combine(seed, OpenMS::hash_int(count));
      }
 
      // Hash the charge
      OpenMS::hash_combine(seed, OpenMS::hash_int(ef.getCharge()));
 
      return seed;
    }
  };
} // namespace std