BayesianProteinInferenceAlgorithm.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
 
//#define INFERENCE_BENCH
 
#include <OpenMS/DATASTRUCTURES/DefaultParamHandler.h>
#include <OpenMS/METADATA/ExperimentalDesign.h>
#include <OpenMS/CONCEPT/ProgressLogger.h>
#include <OpenMS/ML/GRIDSEARCH/GridSearch.h>
 
#include <vector>
#include <functional>
#include <optional>
 
namespace OpenMS
{
  class ConsensusMap;
  namespace Internal
  {
    class IDBoostGraph;
  }
  class PeptideIdentification;
  class ProteinIdentification;
 
  class OPENMS_DLLAPI BayesianProteinInferenceAlgorithm :
      public DefaultParamHandler,
      public ProgressLogger
  {
  public:
    explicit BayesianProteinInferenceAlgorithm(unsigned int debug_lvl = 0);
 
    ~BayesianProteinInferenceAlgorithm() override = default;
 
    void updateMembers_() override;
 
    class GraphInferenceFunctor;
 
    class ExtendedGraphInferenceFunctor;
 
    struct GridSearchEvaluator;
 
 
    void inferPosteriorProbabilities(
        std::vector<ProteinIdentification>& proteinIDs,
        std::vector<PeptideIdentification>& peptideIDs,
        bool greedy_group_resolution,
        std::optional<const ExperimentalDesign> exp_des = std::optional<const ExperimentalDesign>());
 
    void inferPosteriorProbabilities(
        ConsensusMap& cmap,
        bool greedy_group_resolution,
        std::optional<const ExperimentalDesign> exp_des = std::optional<const ExperimentalDesign>());
 
  private:
 
    void inferPosteriorProbabilities_(Internal::IDBoostGraph& ibg);
 
    GridSearch<double,double,double> initGridSearchFromParams_(
        std::vector<double>& alpha_search,
        std::vector<double>& beta_search,
        std::vector<double>& gamma_search
        );
 
    void setScoreTypeAndSettings_(ProteinIdentification& proteinIDs);
 
    // TODO double-check if -1 is maybe the better option
    //  to distinguish between "untouched/unused/unreferenced" (e.g. if somehow
    //  not removed/filtered) and an inferred probability of 0.0. But it might give
    //  problems in FDR algorithms if not ignored/removed correctly
    void resetProteinScores_(ProteinIdentification& protein_id, bool keep_old_as_prior);
 
    std::function<void(PeptideIdentification&/*, const String& run_id*/)> checkConvertAndFilterPepHits_;
 
    unsigned int debug_lvl_;
 
    #ifdef INFERENCE_BENCH
    std::vector<std::pair<double,Size>> debug_times_;
    #endif
 
  };
}