ConfidenceScoring.h

Go to the documentation of this file.

// --------------------------------------------------------------------------
//                   OpenMS -- Open-Source Mass Spectrometry
// --------------------------------------------------------------------------
// Copyright The OpenMS Team -- Eberhard Karls University Tuebingen,
// ETH Zurich, and Freie Universitaet Berlin 2002-2018.
//
// This software is released under a three-clause BSD license:
//  * Redistributions of source code must retain the above copyright
//    notice, this list of conditions and the following disclaimer.
//  * Redistributions in binary form must reproduce the above copyright
//    notice, this list of conditions and the following disclaimer in the
//    documentation and/or other materials provided with the distribution.
//  * Neither the name of any author or any participating institution
//    may be used to endorse or promote products derived from this software
//    without specific prior written permission.
// For a full list of authors, refer to the file AUTHORS.
// --------------------------------------------------------------------------
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
// AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
// ARE DISCLAIMED. IN NO EVENT SHALL ANY OF THE AUTHORS OR THE CONTRIBUTING
// INSTITUTIONS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
// EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// --------------------------------------------------------------------------
// $Maintainer: Hendrik Weisser $
// $Authors: Hannes Roest, Hendrik Weisser $
// --------------------------------------------------------------------------

#pragma once

#include <cmath> // for "exp"
#include <ctime> // for "time" (random number seed)
#include <limits> // for "infinity"
#include <boost/bimap.hpp>
#include <boost/bimap/multiset_of.hpp>
#include <boost/random/uniform_int.hpp>
#include <boost/random/mersenne_twister.hpp>
#include <boost/random/variate_generator.hpp>

#include <OpenMS/FORMAT/FeatureXMLFile.h>
#include <OpenMS/FORMAT/TraMLFile.h>
#include <OpenMS/FORMAT/TransformationXMLFile.h>
#include <OpenMS/KERNEL/FeatureMap.h>

#include "OpenMS/OPENSWATHALGO/ALGO/Scoring.h"
#include <OpenMS/ANALYSIS/MAPMATCHING/TransformationDescription.h>
#include <OpenMS/ANALYSIS/TARGETED/TargetedExperiment.h>

namespace OpenMS
{

  class OPENMS_DLLAPI ConfidenceScoring :
      public ProgressLogger
  {
  public:

      explicit ConfidenceScoring(bool test_mode_ = false) :
        generator_(), rand_gen_(generator_, boost::uniform_int<>())
      {
        if (!test_mode_) rand_gen_.engine().seed(time(nullptr)); // seed with current time
      }

      virtual ~ConfidenceScoring() {}

  protected:

      typedef boost::bimap<double, boost::bimaps::multiset_of<double> > 
      BimapType;

      struct
      {
        double intercept;
        double rt_coef;
        double int_coef;

        double operator()(double diff_rt, double dist_int)
        {
          double lm = intercept + rt_coef * diff_rt * diff_rt + 
            int_coef * dist_int;
          return 1.0 / (1.0 + exp(-lm));
        }
      } glm_;

      struct
      {
        double min_rt;
        double max_rt;
        
        double operator()(double rt)
        {
          return (rt - min_rt) / (max_rt - min_rt) * 100;
        }
      } rt_norm_;

      TargetedExperiment library_; // assay library

      IntList decoy_index_; // indexes of assays to use as decoys

      Size n_decoys_; // number of decoys to use (per feature/true assay)

      Map<String, IntList> transition_map_; // assay (ID) -> transitions (indexes)

      Size n_transitions_; // number of transitions to consider

      TransformationDescription rt_trafo_;

      boost::mt19937 generator_; // random number generation engine

      boost::variate_generator<boost::mt19937&, boost::uniform_int<> > rand_gen_;

      void chooseDecoys_();

      double manhattanDist_(DoubleList x, DoubleList y);

      double getAssayRT_(const TargetedExperiment::Peptide& assay);

      void extractIntensities_(BimapType& intensity_map, Size n_transitions,
                               DoubleList& intensities);

      double scoreAssay_(const TargetedExperiment::Peptide& assay, 
                         double feature_rt, DoubleList& feature_intensities,
                         const std::set<String>& transition_ids = std::set<String>());

      void scoreFeature_(Feature& feature);

  public:

      void initialize(const TargetedExperiment& library, const Size n_decoys, const Size n_transitions, const TransformationDescription& rt_trafo)
      {
        library_ = library; 
        n_decoys_ = n_decoys;
        n_transitions_ = n_transitions;
        rt_trafo_ = rt_trafo;
      }

      void initializeGlm(double intercept, double rt_coef, double int_coef)
      {
        glm_.intercept = intercept;
        glm_.rt_coef = rt_coef;
        glm_.int_coef = int_coef;
      }

      void scoreMap(FeatureMap & features)
      {
        // are there enough assays in the library?
        Size n_assays = library_.getPeptides().size();
        if (n_assays < 2)
        {
          throw Exception::IllegalArgument(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION,
                                           "There need to be at least 2 assays in the library for ConfidenceScoring.");

        }
        if (n_assays - 1 < n_decoys_)
        {
          LOG_WARN << "Warning: Parameter 'decoys' (" << n_decoys_ 
                   << ") is higher than the number of unrelated assays in the "
                   << "library (" << n_assays - 1 << "). "
                   << "Using all unrelated assays as decoys." << std::endl;
        }
        if (n_assays - 1 <= n_decoys_) n_decoys_ = 0; // use all available assays

        decoy_index_.resize(n_assays);
        for (Size i = 0; i < n_assays; ++i) decoy_index_[i] = boost::numeric_cast<Int>(i);

        // build mapping between assays and transitions:
        LOG_DEBUG << "Building transition map..." << std::endl;
        for (Size i = 0; i < library_.getTransitions().size(); ++i)
        {
          const String& ref = library_.getTransitions()[i].getPeptideRef();
          transition_map_[ref].push_back(boost::numeric_cast<Int>(i));
        }
        // find min./max. RT in the library:
        LOG_DEBUG << "Determining retention time range..." << std::endl;
        rt_norm_.min_rt = std::numeric_limits<double>::infinity();
        rt_norm_.max_rt = -std::numeric_limits<double>::infinity();
        for (std::vector<TargetedExperiment::Peptide>::const_iterator it = 
               library_.getPeptides().begin(); it != library_.getPeptides().end();
             ++it)
        {
          double current_rt = getAssayRT_(*it);
          if (current_rt == -1.0) continue; // indicates a missing value
          rt_norm_.min_rt = std::min(rt_norm_.min_rt, current_rt);
          rt_norm_.max_rt = std::max(rt_norm_.max_rt, current_rt);
        }

        // log scoring progress:
        LOG_DEBUG << "Scoring features..." << std::endl;
        startProgress(0, features.size(), "scoring features");

        for (FeatureMap::Iterator feat_it = features.begin(); 
             feat_it != features.end(); ++feat_it)
        {
          LOG_DEBUG << "Feature " << feat_it - features.begin() + 1 
                    << " (ID '" << feat_it->getUniqueId() << "')"<< std::endl;
          scoreFeature_(*feat_it);
          setProgress(feat_it - features.begin());
        }
        endProgress();

      }

  };

}