MRMFeatureFinderScoring.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: Hannes Roest $
// $Authors: Hannes Roest $
// --------------------------------------------------------------------------
 
#pragma once
 
#define USE_SP_INTERFACE
 
// Actual scoring
#include <OpenMS/ANALYSIS/OPENSWATH/OpenSwathScoring.h>
 
#include <OpenMS/ANALYSIS/OPENSWATH/DIAScoring.h>
#include <OpenMS/FEATUREFINDER/EmgScoring.h>
 
// Kernel classes
#include <OpenMS/KERNEL/StandardTypes.h>
#include <OpenMS/ANALYSIS/MAPMATCHING/TransformationDescription.h>
#include <OpenMS/KERNEL/FeatureMap.h>
#include <OpenMS/KERNEL/MRMTransitionGroup.h>
#include <OpenMS/KERNEL/MRMFeature.h>
#include <OpenMS/KERNEL/MSSpectrum.h>
#include <OpenMS/KERNEL/MSChromatogram.h>
#include <OpenMS/ANALYSIS/TARGETED/TargetedExperiment.h>
 
#include <OpenMS/OPENSWATHALGO/DATAACCESS/SwathMap.h>
 
#include <unordered_map>
 
#ifdef _OPENMP
#include <omp.h>
#endif
 
namespace OpenMS
{
  // Forward declaration for optional mobilogram consumer
  class MobilogramParquetConsumer;
  class ProteaseDigestion;
 
 
  class OPENMS_DLLAPI MRMFeatureFinderScoring :
    public DefaultParamHandler,
    public ProgressLogger
  {
 
public:
 
    typedef OpenSwath::LightTransition TransitionType;
    typedef OpenSwath::LightTargetedExperiment TargetedExpType;
    typedef OpenSwath::LightCompound PeptideType;
    typedef OpenSwath::LightProtein ProteinType;
    typedef OpenSwath::LightModification ModificationType;
    // a transition group holds the chromatographic data and peaks across
    // multiple chromatograms from the same compound
    typedef MRMTransitionGroup< MSChromatogram, TransitionType> MRMTransitionGroupType;
    typedef std::map<std::string, MRMTransitionGroupType> TransitionGroupMapType;
 
 
    MRMFeatureFinderScoring();
 
    ~MRMFeatureFinderScoring() override;
 
 
 
    void pickExperiment(const PeakMap & chromatograms,
                        FeatureMap& output,
                        const TargetedExperiment& transition_exp,
                        const TransformationDescription& trafo,
                        const PeakMap& swath_map);
 
    void pickExperiment(const OpenSwath::SpectrumAccessPtr& input,
                        FeatureMap& output,
                        const OpenSwath::LightTargetedExperiment& transition_exp,
                        const TransformationDescription& trafo,
                        const std::vector<OpenSwath::SwathMap>& swath_maps,
                        TransitionGroupMapType& transition_group_map);
 
    static void addProteinIdentificationRun(const OpenSwath::LightTargetedExperiment& transition_exp, FeatureMap& output);
 
    void prepareProteinPeptideMaps_(const OpenSwath::LightTargetedExperiment& transition_exp);
 
  void scorePeakgroups(MRMTransitionGroupType& transition_group,
             const TransformationDescription & trafo,
             const std::vector<OpenSwath::SwathMap>& swath_maps,
             FeatureMap& output,
             bool ms1only = false,
             MobilogramParquetConsumer* mobilogram_consumer = nullptr) const;
 
    void setStrictFlag(bool f)
    {
      strict_ = f;
    }
 
    void setMS1Map(OpenSwath::SpectrumAccessPtr ms1_map)
    {
      ms1_map_ = ms1_map;
    }
 
    void mapExperimentToTransitionList(const OpenSwath::SpectrumAccessPtr& input,
                                       const OpenSwath::LightTargetedExperiment& transition_exp,
                                       TransitionGroupMapType& transition_group_map,
                                       TransformationDescription trafo,
                                       double rt_extraction_window);
private:
 
    struct TransitionGroupCache
    {
      std::vector<double> normalized_library_intensity;
      std::vector<std::string> transition_native_ids;
      std::vector<std::string> precursor_ids;
      std::vector<std::string> transition_native_ids_openms;
      std::vector<std::string> precursor_ids_openms;
    };
 
    struct OpenSwath_Ind_Scores_Pooled : public OpenSwath_Ind_Scores
    {
      void preallocate(Size capacity)
      {
        ind_transition_names.reserve(capacity);
        ind_isotope_correlation.reserve(capacity);
        ind_isotope_overlap.reserve(capacity);
        ind_massdev_score.reserve(capacity);
        ind_xcorr_coelution_score.reserve(capacity);
        ind_xcorr_shape_score.reserve(capacity);
        ind_log_sn_score.reserve(capacity);
        ind_area_intensity.reserve(capacity);
        ind_total_area_intensity.reserve(capacity);
        ind_intensity_score.reserve(capacity);
        ind_apex_intensity.reserve(capacity);
        ind_apex_position.reserve(capacity);
        ind_fwhm.reserve(capacity);
        ind_total_mi.reserve(capacity);
        ind_log_intensity.reserve(capacity);
        ind_intensity_ratio.reserve(capacity);
        ind_mi_ratio.reserve(capacity);
        ind_mi_score.reserve(capacity);
        ind_im_drift.reserve(capacity);
        ind_im_drift_left.reserve(capacity);
        ind_im_drift_right.reserve(capacity);
        ind_im_delta.reserve(capacity);
        ind_im_delta_score.reserve(capacity);
        ind_im_log_intensity.reserve(capacity);
        ind_im_contrast_coelution.reserve(capacity);
        ind_im_contrast_shape.reserve(capacity);
        ind_im_sum_contrast_coelution.reserve(capacity);
        ind_im_sum_contrast_shape.reserve(capacity);
        ind_start_position_at_5.reserve(capacity);
        ind_end_position_at_5.reserve(capacity);
        ind_start_position_at_10.reserve(capacity);
        ind_end_position_at_10.reserve(capacity);
        ind_start_position_at_50.reserve(capacity);
        ind_end_position_at_50.reserve(capacity);
        ind_total_width.reserve(capacity);
        ind_tailing_factor.reserve(capacity);
        ind_asymmetry_factor.reserve(capacity);
        ind_slope_of_baseline.reserve(capacity);
        ind_baseline_delta_2_height.reserve(capacity);
        ind_points_across_baseline.reserve(capacity);
        ind_points_across_half_height.reserve(capacity);
      }
 
      void reset()
      {
        ind_num_transitions = 0;
        ind_transition_names.clear();
        ind_isotope_correlation.clear();
        ind_isotope_overlap.clear();
        ind_massdev_score.clear();
        ind_xcorr_coelution_score.clear();
        ind_xcorr_shape_score.clear();
        ind_log_sn_score.clear();
        ind_area_intensity.clear();
        ind_total_area_intensity.clear();
        ind_intensity_score.clear();
        ind_apex_intensity.clear();
        ind_apex_position.clear();
        ind_fwhm.clear();
        ind_total_mi.clear();
        ind_log_intensity.clear();
        ind_intensity_ratio.clear();
        ind_mi_ratio.clear();
        ind_mi_score.clear();
        ind_im_drift.clear();
        ind_im_drift_left.clear();
        ind_im_drift_right.clear();
        ind_im_delta.clear();
        ind_im_delta_score.clear();
        ind_im_log_intensity.clear();
        ind_im_contrast_coelution.clear();
        ind_im_contrast_shape.clear();
        ind_im_sum_contrast_coelution.clear();
        ind_im_sum_contrast_shape.clear();
        ind_start_position_at_5.clear();
        ind_end_position_at_5.clear();
        ind_start_position_at_10.clear();
        ind_end_position_at_10.clear();
        ind_start_position_at_50.clear();
        ind_end_position_at_50.clear();
        ind_total_width.clear();
        ind_tailing_factor.clear();
        ind_asymmetry_factor.clear();
        ind_slope_of_baseline.clear();
        ind_baseline_delta_2_height.clear();
        ind_points_across_baseline.clear();
        ind_points_across_half_height.clear();
      }
    };
 
    void splitTransitionGroupsDetection_(const MRMTransitionGroupType& transition_group,
                                         MRMTransitionGroupType& transition_group_detection) const;
 
    void splitTransitionGroupsIdentification_(const MRMTransitionGroupType& transition_group,
                                              MRMTransitionGroupType& transition_group_identification,
                                              MRMTransitionGroupType& transition_group_identification_decoy) const;
 
    void scoreIdentification_(MRMTransitionGroupType& transition_group_identification,
                              MRMTransitionGroupType& transition_group_detection,
                              OpenSwathScoring& scorer,
                              const size_t feature_idx,
                              const Int64 feature_id,
                              const std::vector<std::string>& native_ids_detection,
                              const std::vector<OpenSwath::ISignalToNoisePtr>& signal_noise_estimators_identification,
                              const double det_intensity_ratio_score,
                              const double det_mi_ratio_score,
                              const std::vector<OpenSwath::SwathMap>& swath_maps,
                              const double drift_target,
                              RangeMobility& im_range,
                              OpenSwath_Ind_Scores_Pooled& idscores_out,
                              MobilogramParquetConsumer* mobilogram_consumer = nullptr) const;
 
    void prepareScoredFeatureOutput_(OpenMS::MRMFeature& mrmfeature,
                                     const PeptideType& pep,
                                     ProteaseDigestion& pd,
                                     bool swath_present,
                                     double precursor_mz,
                                     bool ms1only) const;
 
    void addScoreMetaValues_(OpenMS::MRMFeature& mrmfeature,
                             const MRMTransitionGroupType& transition_group_detection,
                             const std::vector<OpenSwath::ISignalToNoisePtr>& signal_noise_estimators,
                             const std::vector<OpenSwath::ISignalToNoisePtr>& ms1_signal_noise_estimators,
                             double expected_rt,
                             bool swath_present,
                             bool ms1only) const;
 
    void prepareFeatureOutput_(OpenMS::MRMFeature& mrmfeature, bool ms1only, int charge) const;
 
    void updateMembers_() override;
 
    // parameters
    double rt_extraction_window_;
    double quantification_cutoff_;
    int stop_report_after_feature_;
    bool write_convex_hull_;
    bool strict_;
    bool use_ms1_ion_mobility_;
    bool apply_im_peak_picking_;
    std::string scoring_model_;
    std::string enzyme_;
 
    // scoring parameters
    double rt_normalization_factor_;
    int add_up_spectra_;
    std::string spectrum_addition_method_ ;
    std::string spectrum_merge_method_type_;
    double spacing_for_spectra_resampling_;
    double merge_spectra_by_peak_width_fraction_;
    double uis_threshold_sn_;
    double uis_threshold_peak_area_;
 
    double sn_win_len_;
    unsigned int sn_bin_count_;
    bool write_log_messages_;
 
    double im_extra_drift_;
 
    std::unordered_map<std::string, const PeptideType*> PeptideRefMap_;
    OpenSwath_Scores_Usage su_;
    OpenMS::DIAScoring diascoring_;
    OpenMS::EmgScoring emgscoring_;
 
    // data
    OpenSwath::SpectrumAccessPtr ms1_map_;
 
  };
}
 
#undef run_identifier