NeighborSeq.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, Philipp Wang $
// $Authors: Chris Bielow, Philipp Wang $
// --------------------------------------------------------------------------
 
#pragma once
 
#include <OpenMS/CHEMISTRY/TheoreticalSpectrumGenerator.h>
#include <OpenMS/KERNEL/MSSpectrum.h>
 
#include <vector>
#include <map>
 
 
 
namespace OpenMS
{
    class OPENMS_DLLAPI NeighborSeq
    {
 
    public:
      NeighborSeq(std::vector<AASequence>&& digested_relevant_peptides);
 
      MSSpectrum generateSpectrum(const AASequence& peptide_sequence);
 
      static bool isNeighborSpectrum(const MSSpectrum& spec1, const MSSpectrum& spec2, const double min_shared_ion_fraction, const double mz_bin_size);
      static int computeSharedIonCount(const MSSpectrum& spec1, const MSSpectrum& spec2, const double& mz_bin_size);
 
      bool isNeighborPeptide(const AASequence& neighbor_candidate,
                             const double mass_tolerance_pc,
                             const bool mass_tolerance_pc_ppm,
                             const double min_shared_ion_fraction,
                             const double mz_bin_size);
 
      struct NeighborStats
      {
        int unfindable_peptides = 0; 
        int findable_no_neighbors = 0; 
        int findable_one_neighbor = 0; 
        int findable_multiple_neighbors = 0; 
        
        int total() const
        {
          return unfindable_peptides + findable_no_neighbors + findable_one_neighbor + findable_multiple_neighbors;
        }
 
        int percentOfTotal_(int count) const
        {
          const int t = total();
          return (t == 0) ? 0 : count * 100 / t;
        }
 
        std::string unfindable() const
        {
          return StringUtils::toStr(unfindable_peptides) + " (" + StringUtils::toStr(percentOfTotal_(unfindable_peptides)) + "%)";
        }
 
        std::string noNB() const
        {
          return StringUtils::toStr(findable_no_neighbors) + " (" + StringUtils::toStr(percentOfTotal_(findable_no_neighbors)) + "%)";
        }
 
        std::string oneNB() const
        {
          return StringUtils::toStr(findable_one_neighbor) + " (" + StringUtils::toStr(percentOfTotal_(findable_one_neighbor)) + "%)";
        }
 
        std::string multiNB() const
        {
          return StringUtils::toStr(findable_multiple_neighbors) + " (" + StringUtils::toStr(percentOfTotal_(findable_multiple_neighbors)) + "%)";
        }
      };
 
      NeighborStats getNeighborStats() const;
 
    protected:
      std::map<double, std::vector<int>> createMassLookup_();
      
      auto findCandidatePositions_(const double mono_weight, double mass_tolerance, const bool mass_tolerance_pc_ppm);
 
 
    private:
      const std::vector<AASequence>& digested_relevant_peptides_; 
      std::map<double, std::vector<int>> mass_position_map_; 
 
      TheoreticalSpectrumGenerator spec_gen_; 
      const Residue* x_residue_; 
 
      std::vector<int> neighbor_stats_; 
 
  }; // class NeighborSeq
 
} // namespace OpenMS