PeptideSearchEngineFIAlgorithm.h

Go to the documentation of this file.

// Copyright (c) 2002-present, The OpenMS Team -- EKU Tuebingen, ETH Zurich, and FU Berlin
// SPDX-License-Identifier: BSD-3-Clause
//
// --------------------------------------------------------------------------
// $Maintainer:  $
// $Authors: Raphael Förster $
// --------------------------------------------------------------------------
 
#pragma once
 
#include <OpenMS/CONCEPT/ProgressLogger.h>
#include <OpenMS/DATASTRUCTURES/DefaultParamHandler.h>
 
#include <OpenMS/CHEMISTRY/ModifiedPeptideGenerator.h>
#include <OpenMS/FORMAT/FASTAFile.h>
#include <OpenMS/KERNEL/MSExperiment.h>
#include <OpenMS/DATASTRUCTURES/StringView.h>
#include <OpenMS/METADATA/PeptideIdentificationList.h>
#include <OpenMS/ANALYSIS/ID/OpenSearchModificationAnalysis.h>
 
#include <vector>
 
namespace OpenMS
{
 
class OPENMS_DLLAPI PeptideSearchEngineFIAlgorithm :
  public DefaultParamHandler,
  public ProgressLogger
{
  public:
    PeptideSearchEngineFIAlgorithm(); 
 
    enum class ExitCodes
    {
      EXECUTION_OK,
      INPUT_FILE_EMPTY,
      UNEXPECTED_RESULT,
      UNKNOWN_ERROR,
      ILLEGAL_PARAMETERS
    };
 
    struct SearchResult
    {
      ExitCodes exit_code = ExitCodes::EXECUTION_OK;
      std::vector<ProteinIdentification> protein_ids;
      PeptideIdentificationList peptide_ids;
      OpenSearchModificationAnalysis::OpenSearchAnalysisResult modification_analysis;
      bool is_open_search = false;
    };
 
    ExitCodes search(const String& in_mzML,
      const String& in_db,
      std::vector<ProteinIdentification>& prot_ids,
      PeptideIdentificationList& pep_ids) const;
 
    SearchResult searchWithModificationAnalysis(const String& in_mzML,
                                                const String& in_db,
                                                const String& output_base_name = "") const;
 
    ExitCodes search(PeakMap& spectra,
                     const std::vector<FASTAFile::FASTAEntry>& fasta_db,
                     std::vector<ProteinIdentification>& prot_ids,
                     PeptideIdentificationList& pep_ids) const;
 
    SearchResult searchWithModificationAnalysis(PeakMap& spectra,
                                                const std::vector<FASTAFile::FASTAEntry>& fasta_db,
                                                const String& output_base_name = "") const;
 
  protected:
    void updateMembers_() override;
 
    struct AnnotatedHit_
    {
      AASequence sequence;
      /*
      StringView sequence;
      SignedSize peptide_mod_index; ///< enumeration index of the non-RNA peptide modification
      */
      double score = 0; 
      std::vector<PeptideHit::PeakAnnotation> fragment_annotations;
      double prefix_fraction = 0; 
      double suffix_fraction = 0; 
      double mean_error = 0.0; 
      int isotope_error = 0; 
      uint16_t applied_charge = 0; 
      double delta_mass = 0.0; 
 
      static bool hasBetterScore(const AnnotatedHit_& a, const AnnotatedHit_& b)
      {
        if (a.score != b.score) return a.score > b.score;
        return a.sequence < b.sequence;
      }
    };
 
    static void preprocessSpectra_(PeakMap& exp, double fragment_mass_tolerance, bool fragment_mass_tolerance_unit_ppm);
 
    void postProcessHits_(const PeakMap& exp,
      std::vector<std::vector<PeptideSearchEngineFIAlgorithm::AnnotatedHit_> >& annotated_hits,
      std::vector<ProteinIdentification>& protein_ids,
      PeptideIdentificationList& peptide_ids,
      Size top_hits,
      const StringList& modifications_fixed,
      const StringList& modifications_variable,
      Int peptide_missed_cleavages,
      double precursor_mass_tolerance,
      double fragment_mass_tolerance,
      const String& precursor_mass_tolerance_unit_ppm,
      const String& fragment_mass_tolerance_unit_ppm,
      const Int precursor_min_charge,
      const Int precursor_max_charge,
      const String& enzyme,
      const String& database_name) const;
 
    double precursor_mass_tolerance_;
    String precursor_mass_tolerance_unit_;
 
    Size precursor_min_charge_;
    Size precursor_max_charge_;
 
    IntList precursor_isotopes_;
 
    double fragment_mass_tolerance_;
 
    String fragment_mass_tolerance_unit_;
 
    StringList modifications_fixed_;
 
    StringList modifications_variable_;
 
    Size modifications_max_variable_mods_per_peptide_;
 
    String enzyme_;
 
    bool decoys_;
 
    StringList annotate_psm_;
 
    Size peptide_min_size_;
    Size peptide_max_size_;
    Size peptide_missed_cleavages_;
 
    String peptide_motif_;
 
    Size report_top_hits_;
 
    void logModificationAnalysisSummary_(const SearchResult& result,
                                         const String& output_base_name) const;
 
    bool isOpenSearchMode_() const
    {
      return precursor_mass_tolerance_unit_ == "ppm"
               ? (precursor_mass_tolerance_ > 1000.0)
               : (precursor_mass_tolerance_ > 1.0);
    }
};
 
} // namespace