NASequence.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: Samuel Wein $
// $Authors: Samuel Wein, Timo Sachsenberg, Hendrik Weisser $
// --------------------------------------------------------------------------
 
#pragma once
 
#include <OpenMS/CHEMISTRY/EmpiricalFormula.h>
#include <OpenMS/CHEMISTRY/Ribonucleotide.h>
#include <OpenMS/CONCEPT/HashUtils.h>
#include <OpenMS/CONCEPT/Types.h>
#include <OpenMS/DATASTRUCTURES/StringUtils.h>
#include <OpenMS/MATH/MathFunctions.h>
#include <functional>
#include <iosfwd>
#include <vector>
 
namespace OpenMS
{
  class OPENMS_DLLAPI NASequence
  {
  public:
    enum NASFragmentType
    {                //< NB: Not all fragments types are valid for all residue types, this class should probably get split
      Full = 0,      
      Internal,      
      FivePrime,     
      ThreePrime,    
      AIon,          
      BIon,          
      CIon,          
      XIon,          
      YIon,          
      ZIon,          
      Precursor,     
      BIonMinusH20,  
      YIonMinusH20,  
      BIonMinusNH3,  
      YIonMinusNH3,  
      NonIdentified, 
      Unannotated,   
      WIon,          
      AminusB,       
      DIon,          
      SizeOfNASFragmentType
    };
 
    using ConstRibonucleotidePtr = const Ribonucleotide*;
 
    class Iterator;
 
    class OPENMS_DLLAPI ConstIterator
    {
    public:
      typedef Ribonucleotide value_type;
      typedef const value_type& const_reference;
      typedef value_type& reference;
      typedef const value_type* const_pointer;
      typedef std::vector<const value_type*>::difference_type difference_type;
      typedef const value_type* pointer;
      typedef std::random_access_iterator_tag iterator_category;
 
      ConstIterator() = default;
 
      ConstIterator(const std::vector<const Ribonucleotide*>* vec_ptr, difference_type position)
      {
        vector_ = vec_ptr;
        position_ = position;
      }
 
      ConstIterator(const ConstIterator& rhs) : vector_(rhs.vector_), position_(rhs.position_)
      {
      }
 
      ConstIterator(const NASequence::Iterator& rhs) : vector_(rhs.vector_), position_(rhs.position_)
      {
      }
 
      virtual ~ConstIterator()
      {
      }
 
 
      ConstIterator& operator=(const ConstIterator& rhs)
      {
        if (this != &rhs)
        {
          position_ = rhs.position_;
          vector_ = rhs.vector_;
        }
        return *this;
      }
 
      const_reference operator*() const
      {
        return *(*vector_)[position_];
      }
 
      const_pointer operator->() const
      {
        return (*vector_)[position_];
      }
 
      const ConstIterator operator+(difference_type diff) const
      {
        return ConstIterator(vector_, position_ + diff);
      }
 
      difference_type operator-(ConstIterator rhs) const
      {
        return position_ - rhs.position_;
      }
 
      const ConstIterator operator-(difference_type diff) const
      {
        return ConstIterator(vector_, position_ - diff);
      }
 
      bool operator==(const ConstIterator& rhs) const
      {
        return (std::tie(vector_, position_) == std::tie(rhs.vector_, rhs.position_));
      }
 
      bool operator!=(const ConstIterator& rhs) const
      {
        return !(operator==(rhs));
      }
 
      ConstIterator& operator++()
      {
        ++position_;
        return *this;
      }
 
      ConstIterator& operator--()
      {
        --position_;
        return *this;
      }
 
 
    protected:
      // pointer to the vector
      const std::vector<const Ribonucleotide*>* vector_;
 
      // position in the vector
      difference_type position_;
    };
 
 
    class OPENMS_DLLAPI Iterator
    {
    public:
      friend class NASequence::ConstIterator;
 
      typedef Ribonucleotide value_type;
      typedef const value_type& const_reference;
      typedef value_type& reference;
      typedef const value_type* const_pointer;
      typedef const value_type* pointer;
      typedef std::vector<const value_type*>::difference_type difference_type;
 
      Iterator() = default;
 
      Iterator(std::vector<const Ribonucleotide*>* vec_ptr, difference_type position)
      {
        vector_ = vec_ptr;
        position_ = position;
      }
 
      Iterator(const Iterator& rhs) : vector_(rhs.vector_), position_(rhs.position_)
      {
      }
 
      virtual ~Iterator()
      {
      }
 
 
      Iterator& operator=(const Iterator& rhs)
      {
        if (this != &rhs)
        {
          position_ = rhs.position_;
          vector_ = rhs.vector_;
        }
        return *this;
      }
 
      const_reference operator*() const
      {
        return *(*vector_)[position_];
      }
 
      const_pointer operator->() const
      {
        return (*vector_)[position_];
      }
 
      pointer operator->()
      {
        return (*vector_)[position_];
      }
 
      const Iterator operator+(difference_type diff) const
      {
        return Iterator(vector_, position_ + diff);
      }
 
      difference_type operator-(Iterator rhs) const
      {
        return position_ - rhs.position_;
      }
 
      const Iterator operator-(difference_type diff) const
      {
        return Iterator(vector_, position_ - diff);
      }
 
      bool operator==(const Iterator& rhs) const
      {
        return (std::tie(vector_, position_) == std::tie(rhs.vector_, rhs.position_));
      }
 
      bool operator!=(const Iterator& rhs) const
      {
        return !this->operator==(rhs);
      }
 
      Iterator& operator++()
      {
        ++position_;
        return *this;
      }
 
      Iterator& operator--()
      {
        --position_;
        return *this;
      }
 
 
    protected:
      std::vector<const Ribonucleotide*>* vector_;
 
      // position in the vector
      difference_type position_;
    };
 
  public:
    /*
     * Default constructors and assignment operators.
     */
    NASequence() = default;                               
    NASequence(const NASequence&) = default;              
    NASequence(NASequence&&) = default;                   
    NASequence& operator=(const NASequence&) & = default; 
    NASequence& operator=(NASequence&&) & = default;      
 
    NASequence(std::vector<const Ribonucleotide*> s, const RibonucleotideChainEnd* five_prime, const RibonucleotideChainEnd* three_prime);
 
    virtual ~NASequence() = default; 
 
    bool operator==(const NASequence& rhs) const; 
    bool operator!=(const NASequence& rhs) const; 
    bool operator<(const NASequence& rhs) const;  
 
    void setSequence(const std::vector<const Ribonucleotide*>& seq);
 
    const std::vector<const Ribonucleotide*>& getSequence() const
    {
      return seq_;
    }
 
    std::vector<const Ribonucleotide*>& getSequence()
    {
      return seq_;
    }
 
    void set(size_t index, const Ribonucleotide* r);
 
    const Ribonucleotide* get(size_t index)
    {
      return seq_[index];
    }
 
    inline const Ribonucleotide*& operator[](size_t index)
    {
      return seq_[index];
    }
 
    inline const Ribonucleotide* const& operator[](size_t index) const
    {
      return seq_[index];
    }
 
    bool empty() const;
    size_t size() const;
    void clear();
 
    bool hasFivePrimeMod() const;
    void setFivePrimeMod(const RibonucleotideChainEnd* r);
    const RibonucleotideChainEnd* getFivePrimeMod() const;
    bool hasThreePrimeMod() const;
    void setThreePrimeMod(const RibonucleotideChainEnd* r);
    const RibonucleotideChainEnd* getThreePrimeMod() const;
 
    inline Iterator begin()
    {
      return Iterator(&seq_, 0);
    }
 
    inline ConstIterator begin() const
    {
      return ConstIterator(&seq_, 0);
    }
 
    inline Iterator end()
    {
      return Iterator(&seq_, (Int)seq_.size());
    }
 
    inline ConstIterator end() const
    {
      return ConstIterator(&seq_, (Int)seq_.size());
    }
 
    inline ConstIterator cbegin() const
    {
      return ConstIterator(&seq_, 0);
    }
 
    inline ConstIterator cend() const
    {
      return ConstIterator(&seq_, (Int)seq_.size());
    }
 
 
    double getMonoWeight(NASFragmentType type = Full, Int charge = 0) const;
 
    double getAverageWeight(NASFragmentType type = Full, Int charge = 0) const;
 
    EmpiricalFormula getFormula(NASFragmentType type = Full, Int charge = 0) const;
 
    NASequence getPrefix(Size length) const;
 
    NASequence getSuffix(Size length) const;
 
    NASequence getSubsequence(Size start = 0, Size length = Size(-1)) const;
 
    NASequence getReversed(bool keep_five_prime = false, bool keep_three_prime = false) const;
 
    NASequence getShuffled(Math::RandomShuffler& shuffler, double identity_threshold = 0.5,
                           int max_attempts = 30,
                           bool keep_five_prime = false, bool keep_three_prime = false) const;
 
    float getSequenceIdentity(const NASequence& other) const;
 
    static NASequence fromString(const std::string& s);
 
    friend OPENMS_DLLAPI std::ostream& operator<<(std::ostream& os, const NASequence& seq);
 
    static NASequence fromString(const char* s);
 
    std::string toString() const;
 
  private:
    // TODO: query RNA / DNA depending on type
    static void parseString_(const std::string& s, NASequence& nas);
 
    // TODO: query RNA / DNA depending on type
    static std::string::const_iterator parseMod_(const std::string::const_iterator str_it, const std::string& str, NASequence& nas);
 
    std::vector<const Ribonucleotide*> seq_;
 
    const RibonucleotideChainEnd* five_prime_ = nullptr;
    const RibonucleotideChainEnd* three_prime_ = nullptr;
  };
 
} // namespace OpenMS
 
// Hash function specialization for NASequence
// Placed in std namespace to allow use with std::unordered_map/set
namespace std
{
  template<>
  struct hash<OpenMS::NASequence>
  {
    std::size_t operator()(const OpenMS::NASequence& seq) const noexcept
    {
      std::size_t seed = 0;
 
      // Hash each ribonucleotide in the sequence
      for (const auto& ribo : seq)
      {
        // Hash the code (stable identifier)
        const std::string& code = ribo.getCode();
        OpenMS::hash_combine(seed, OpenMS::fnv1a_hash_string(code));
      }
 
      // Hash 5' terminal modification if present
      const OpenMS::RibonucleotideChainEnd* five_prime = seq.getFivePrimeMod();
      if (five_prime != nullptr)
      {
        // Use a different seed offset for 5' to distinguish from 3'
        std::size_t five_hash = OpenMS::fnv1a_hash_string(five_prime->getCode());
        OpenMS::hash_combine(seed, five_hash ^ 0x355052494dULL); // "5PRIM" in hex-like
      }
 
      // Hash 3' terminal modification if present
      const OpenMS::RibonucleotideChainEnd* three_prime = seq.getThreePrimeMod();
      if (three_prime != nullptr)
      {
        // Use a different seed offset for 3'
        std::size_t three_hash = OpenMS::fnv1a_hash_string(three_prime->getCode());
        OpenMS::hash_combine(seed, three_hash ^ 0x335052494dULL); // "3PRIM" in hex-like
      }
 
      return seed;
    }
  };
} // namespace std