IDMergerAlgorithm.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: Julianus Pfeuffer $
// $Authors: Julianus Pfeuffer $
// --------------------------------------------------------------------------
 
#pragma once
 
#include <OpenMS/DATASTRUCTURES/DefaultParamHandler.h>
#include <OpenMS/CONCEPT/ProgressLogger.h>
#include <OpenMS/KERNEL/ConsensusMap.h>
#include <OpenMS/METADATA/ExperimentalDesign.h>
#include <OpenMS/METADATA/ProteinIdentification.h>
#include <OpenMS/METADATA/PeptideIdentificationList.h>
 
#include <map>
#include <unordered_set>
 
namespace OpenMS
{
 
  //TODO add params for checking consistency (i.e. how strict to check)
  //TODO add another subclass that does score-aware merging? (i.e. only keep best per peptide[sequence])
 
  class OPENMS_DLLAPI IDMergerAlgorithm:
    public DefaultParamHandler,
    public ProgressLogger
  {
  public:
    explicit IDMergerAlgorithm (const std::string& runIdentifier = "merged", bool addTimeStampToID = true);
 
    void insertRuns(std::vector<ProteinIdentification>&& prots,   
                    PeptideIdentificationList&& peps);
                    
    void insertRuns(const std::vector<ProteinIdentification>& prots,
                    const PeptideIdentificationList& peps);
 
    //TODO add methods to just insert prots or just peps. Especially makes sense if you do re-indexing anyway,
    // then you do not need the proteins. But then we need origin information. Either externally in form of a
    // std::string or StringList (like the one from ProteinID.getPrimaryMSRunPath). Or by having the file annotated
    // at the PeptideID (with getBasename maybe?)
    // Current solution would be to clear the ProteinIdentification if you do not need the proteins and add all the
    // necessary information about origin(s) to this ProteinIdentification.
 
    void returnResultsAndClear(ProteinIdentification& prots,
                   PeptideIdentificationList& peps);
 
  private:
 
    std::string getNewIdentifier_(bool addTimeStampToID) const;
 
    static void copySearchParams_(const ProteinIdentification& from, ProteinIdentification& to);
 
    bool checkOldRunConsistency_(
        const std::vector<ProteinIdentification>& protRuns,
        const std::string& experiment_type) const;
 
    bool checkOldRunConsistency_(
        const std::vector<ProteinIdentification>& protRuns,
        const ProteinIdentification& ref,
        const std::string& experiment_type) const;
 
    void insertProteinIDs_(
        std::vector<ProteinIdentification>&& old_protRuns
    );
 
    void updateAndMovePepIDs_(
        PeptideIdentificationList&& pepIDs,
        const std::map<std::string, Size>& runID_to_runIdx,
        const std::vector<StringList>& originFiles,
        bool annotate_origin
    );
 
    void movePepIDsAndRefProteinsToResultFaster_(
        PeptideIdentificationList&& pepIDs,
        std::vector<ProteinIdentification>&& old_protRuns
    );
 
    ProteinIdentification prot_result_;
 
    PeptideIdentificationList pep_result_;
 
    static size_t accessionHash_(const ProteinHit& p){
      return std::hash<std::string>()(p.getAccession());
    }
    
    static bool accessionEqual_(const ProteinHit& p1, const ProteinHit& p2){
      return p1.getAccession() == p2.getAccession();
    }
    
    using hash_type = std::size_t (*)(const ProteinHit&);
    
    using equal_type = bool (*)(const ProteinHit&, const ProteinHit&);
    
    std::unordered_set<ProteinHit, hash_type, equal_type> collected_protein_hits_;
 
    bool filled_ = false;
 
    std::map<std::string, Size> file_origin_to_idx_;
 
    std::string id_;
 
    bool fixed_identifier_;
  };
} // namespace OpenMS